KR20010028767A - Channel equalizer and channel equalization method - Google Patents

Abstract

The present invention relates to a demodulation system of a modulation method having a reference signal, and more particularly, to a channel equalization method and a channel equalizer capable of performing high-speed channel equalization using a G-PSEUDO algorithm. To this end, the present invention calculates a filter tap coefficient for removing a multipath by receiving an error signal, and is updated with the calculated coefficient to remove the multipath of the received signal sequence, and an interval in which the reference signal is transmitted. The difference between the known reference signal and the received signal is output as the error signal, and in the section where the reference signal is not transmitted, the error signal of the received signal is obtained by decision-directed technique and blind equalization technique, respectively, and the detected VSB The channel equalizer is configured by an error calculator which calculates an error signal by calculating the parameters variable according to the mode and outputs the error signal to the adaptive filter. As a result, channel equalization may not diverge even when channel conditions are switched or non-ideal signals are received in accordance with the transition from blind equalization to decision-oriented. There is a characteristic.

Description

CHANNEL EQUALIZER AND CHANNEL EQUALIZATION METHOD}

The present invention relates to a demodulation system of a modulation method having a reference signal, and more particularly, to a channel equalizer and a channel equalization method for calculating an error signal for compensating for a transmission channel error using a G_PSEUDO algorithm.

The first consideration in constructing a digital communication system relates to a channel environment in which communication is performed. Generally considered channel environment includes cable, dominant, ground, etc. Each channel environment has its own characteristics. Among these, the terrestrial channel exhibits the worst channel characteristic. In general, when the communication is performed through the terrestrial channel, many errors occur in the received signal. The most common cause of such channel errors is the multipath characteristics caused by reflectors that interfere with the radio wave transmission process. This impedes the demodulation and decoding process of the receiver by transmitting the transmitted radio through several paths in addition to being delivered directly to the receiver. In communication systems, receivers employ channel equalizers to compensate for errors on these channels.

The channel equalizer is a kind of digital filter and has a reverse response to the impulse response of a channel having a multipath characteristic. However, since the characteristics of the transmission channel vary from region to region, and the characteristic varies with time in the region, the channel equalizer is designed as an adaptive filter that adaptively updates coefficients rather than a filter having fixed coefficients. The adaptive update of the filter coefficients is made to obtain and minimize the difference (ie, error signal) between the transmitted signal and the received signal. Representative examples of methods for obtaining the error signal as described above in the channel equalizer are as follows.

First, there is a method of using a training sequence that is a reference signal. The training signal sequence is a predetermined signal transmitted from the transmitter for channel equalization. The receiver extracts the channel error signal from the reception state of the signal and updates the coefficient of the channel equalizer based on the extracted channel error signal. Although the channel error signal can be reliably obtained using the training signal sequence, in general, the training signal sequence is a kind of redundancy that does not include information and thus cannot be transmitted frequently. In the case of ATSC standard, it is repeatedly transmitted at about 78usec time length every 24msec.

As a second method of obtaining an error signal, there is a decision directed technique. The decision directing technique determines the received signal as an adjacent symbol value in order to extract a channel error signal in a section in which the training signal sequence is not transmitted. This technique is the simplest, but when there are many channel errors, there is a disadvantage that the decision itself causes an error in the channel error signal.

As a third method of obtaining an error signal, there is a blind equalization technique. The blind equalization technique uses the statistical characteristics of the transmitted signal to overcome the shortcomings of the decision-oriented technique, which can be applied even when there are many channel errors. However, since the jitter value is large after convergence, the channel alone is used. It is difficult to complete equalization.

As described above, each method of obtaining an error signal in the channel equalizer cannot be used separately because it has its own characteristics. Therefore, in a general channel equalizer, an error signal generated by a training signal sequence is obtained to update an adaptive filter coefficient in a section where a training signal sequence is transmitted, and a coefficient of the adaptive filter is first determined using a blind equalization technique in a section other than the training signal sequence. After some updates and convergence, channel equalization is completed using decision-oriented techniques.

In this case, it is not a problem because the difference between the transmitted signal and the received signal can be accurately calculated in the section in which the training signal sequence is transmitted. There is a disadvantage that is made discontinuously, thereby causing a problem that the channel equalization completion time is delayed. In addition, when the channel state is switched or a non-ideal signal is received in accordance with the time point of switching from the blind equalization technique to the decision-oriented technique, a problem may occur.

Accordingly, an object of the present invention is to provide a channel equalizer capable of minimizing divergence that may occur in a section other than the training signal sequence that is the reference signal in a demodulation system having a reference signal.

It is still another object of the present invention to provide a channel equalization method and a channel equalizer capable of performing fast channel equalization using a G-PSEUDO algorithm in a demodulation system having a reference signal.

The present invention for achieving the above object in the channel equalizer constituting a demodulation system of the modulation method having a reference signal,

An adaptive filter unit which receives an error signal, calculates a filter tap coefficient for removing the multipath, and updates the calculated coefficient to remove the multipath of the received signal sequence;

In the section in which the reference signal is transmitted, the difference between the known reference signal and the received signal is output as the error signal, and in the section in which the reference signal is not transmitted, the error of the received signal is determined by a decision-oriented technique and a blind equalization technique. And an error calculator which calculates an error signal by calculating the signals and calculates the error signals by varying the parameters according to the detected VSB mode.

The present invention also provides a channel equalization method in a broadcast receiving system for removing a multipath ghost from a received signal including a reference signal.

A first process of updating a coefficient of a filter for removing the multipath ghost by obtaining a difference between a known reference signal and a received signal in an interval in which the reference signal is transmitted, and setting the difference as an error signal;

In the section in which the reference signal is not transmitted, error signals of the received signals are obtained by decision-directed techniques and blind equalization techniques, respectively, and they are calculated by calculating parameters according to the VSB mode detected and G-PSEUDO algorithm. And a second process of updating the coefficient of the filter with an error signal.

1 is an exemplary channel equalizer configuration for explaining the operation of the channel equalizer.

2 is a detailed block diagram of a channel equalizer according to an embodiment of the present invention.

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

First, before explaining the configuration and operation of the channel equalizer according to the embodiment of the present invention, an example of the configuration of the channel equalizer employed in the HDTV receiver is illustrated in FIG. 1. The channel equalizer of the VSB receiver shown in FIG. 1 schematically shows the diagram described on page 108 of the book "Guide to the Use of the Digital Television Standard For HDTV Transmission" issued by ATSC.

The general channel equalizer 100 as shown in FIG. 1 includes an error calculator 120 that calculates a difference from a known reference signal or a determination signal from an output of the adaptive filter means 110 for removing a multipath. And an adaptive filter unit configured to receive an error signal e (i) from the error calculator 120 and update a filter tap coefficient to remove the multipath, and a filter to remove the multipath using the updated coefficient. 110). On the other hand, the selection of the reference signal (refer to the training signal sequence) and the determination signal is made by a timing control signal, which is generated by a field sync detector (not shown). The timing control signal refers to a signal that is activated when field synchronization is detected so that the channel equalizer 100 can recognize a training signal sequence corresponding to a reference signal in a transmission signal format.

Hereinafter, the configuration and operation of the channel equalizer according to the embodiment of the present invention which are different from the configuration of the general channel equalizer described above will be described.

2 shows a block diagram of a channel equalizer and its surroundings in accordance with an embodiment of the present invention. Referring to FIG. 2, the channel equalizer according to the embodiment of the present invention is composed of an adaptive filter unit 110 and an error calculating unit 120 using a G-PSEUDO algorithm. The VSB mode detector and the field sync detector are positioned to support the operation of the error calculator 120.

First, the adaptive filter unit 110 receives the error signal e (i) output from the error calculator 120 and calculates a filter coefficient calculator 114 for calculating a filter tap coefficient for removing the multipath, and the filter. The filter 112 is updated with coefficients calculated by the coefficient calculating unit 114 to remove the multipath of the received signal sequence.

On the other hand, the error calculator 120 decides the received signal y (i) as a near symbol value in a section in which a training signal sequence as a reference signal is not transmitted, and determines a decision-oriented error signal. And a decision oriented error signal generation unit 121 for generating and outputting the signal. The decision oriented error signal generation unit 121 includes a determiner 122 and an adder 124. The decision-directed error signal output from the adder 124 When formulated as follows.

In addition, the error calculation unit 120 uses a blind error signal using statistical characteristics of a transmission signal in a section in which the training signal sequence as the reference signal is not transmitted. And a blind error signal generation unit 125 for generating and outputting the blind error signal generation unit 125. The blind error signal generation unit 125 includes a blind equalizer 126 and an adder 128. The blind error signal output from the adder 128 When formulated, it can be expressed as Equation 2 below.

In addition, the error calculator 120 is provided with an error combiner 130, the error combiner 130 is blind with the decision-oriented error signal generator 121 in accordance with the mode detection signal output from the VSB mode detector Error signals output from the error signal generator 125 ( , ) Is computed based on the G-PSEUDO algorithm. The error signal processed in this way Is output as an error signal in a random signal section. The G-PSEUDO algorithm may be expressed by Equation 3 below, and it is assumed that the random signal section refers to a section other than a training signal sequence that is a reference signal.

In Equation 3 Is a parameter that combines error values, and the performance of the G-PSEUDO algorithm depends on these parameters. In the embodiment of the present invention Is variably set at a constant rate according to the VSB mode. For reference, 2, 4, 8, 15, and 16 VSB modes exist in the VSB mode, and a signal used for VSB mode detection is performed by using a VSB mode signal included in a field sync segment repeated every 313 segments. Detect. In the embodiment of the present invention, the parameters in 8VSB, 16VSB mode Is assumed to be set at a ratio of 4: 1, and in the case of 15 VSB mode, Suppose is set at a ratio of 1: 2. The reason for this assumption is that in the VSB mode in the random signal section, This is because the most stable channel equalization characteristic is shown when is set.

The error calculator 120 includes a reference signal generator 132 and an adder 134. The reference signal generator 132 generates a reference signal having a predetermined pattern such as a PN sequence (Pseudo Noise Sequence). In the adder 134, the difference between the output signal y (i) of the filter 112 and the reference signal is an error signal. Is output. In the following description, the reference signal generator 132 and the adder 134 will be referred to as an error signal generator in the reference signal section.

A MUX 136 is provided at the output terminal of the error calculator 120. The MUX 136 is an error output from the error combiner 130 or the adder 134 according to a timing control signal output from a field sink detector. One of the signals is selectively output to the adaptive filter unit 110.

According to the configuration as described above, the present invention, if the training signal sequence section, such as the reference signal is first detected by the field sync detector to control the MUX 136 accordingly obtained from the difference between the received signal y (i) and the reference signal Error signal The channel equalization is performed by updating the coefficients of the filter tap using. In the random signal section in which the timing control signal is disabled, the error signal output from the error combiner 130 is used. The channel equalization is performed by updating the coefficients of the filter tap using.

Therefore, unlike the prior art in which channel equalization is discontinuously performed by using blind equalization or decision-directed techniques in random signal sections, the present invention can perform continuous channel equalization using G-PSEUDO algorithm even in random signal sections. have.

As described above, the present invention is different from the prior art in which channel equalization is discontinuously performed by using blind equalization or decision-directed techniques in random signal intervals, and thus, continuous channel equalization is performed using G-PSEUDO algorithm even in random signal intervals. Because it can be performed, there is an advantage of shortening the VSB channel equalization completion time,

In line with the transition from blind equalization to decision-oriented, there is no possibility that channel equalization can diverge even when channel conditions are switched or non-ideal signals are received. have.

Claims (7)

In the channel equalizer constituting the demodulation system of the modulation method having a reference signal, An adaptive filter unit which receives an error signal, calculates a filter tap coefficient for removing the multipath, and updates the calculated coefficient to remove the multipath of the received signal sequence; In the section in which the reference signal is transmitted, the difference between the known reference signal and the received signal is output as the error signal, and in the section in which the reference signal is not transmitted, the error of the received signal is determined by a decision-oriented technique and a blind equalization technique, respectively. And an error calculator which calculates an error signal by calculating the signals and calculates the error signals by varying the parameters according to the detected VSB mode. The apparatus of claim 1, wherein the error calculator comprises: an error calculator; In the period in which the reference signal is not transmitted, the received signal is determined as an adjacent symbol value, and a difference between the decision value and the received signal is determined by an error-directed error signal ( A decision-oriented error signal generator that generates and outputs In the period in which the reference signal is not transmitted, the blind error signal ( Blind error signal generator for generating and outputting According to the VSB mode detection signal, the decision-directed error signal ( ) And blind error signal ( ) Is processed according to the G-PSEUDO algorithm as shown in Equation 4 below, whereby the first error signal ( An error combiner to output An error signal generator in a reference signal section for outputting a difference between the reference signal and the received signal known in advance as a second error signal; And a multiplexer configured to selectively output one of the first error signal and the second error signal as the error signal according to a timing control signal that is activated for a predetermined time when detecting field sync. The method according to claim 2, which is used as a parameter in the G-PSEUDO algorithm. The channel equalizer, characterized in that the ratio is set to 4: 1 in 8VSB, 16VSB mode, respectively. The method according to claim 2, which is used as a parameter in the G-PSEUDO algorithm. Are each set at a ratio of 1: 2 in 15 VSB mode. A channel equalization method in a broadcast receiving system for removing a multipath ghost from a received signal including a reference signal, A first process of updating a coefficient of a filter for removing the multipath ghost by obtaining a difference between a known reference signal and a received signal in an interval in which the reference signal is transmitted, and setting the difference as an error signal; In the section in which the reference signal is not transmitted, error signals of the received signals are obtained by decision-directed techniques and blind equalization techniques, respectively, and they are calculated by calculating parameters according to the VSB mode detected and G-PSEUDO algorithm. And a second step of updating the coefficient of the filter with an error signal. 6. The method of claim 5, wherein the parameter is used in the G-PSEUDO algorithm. The channel equalization method is characterized in that the ratio is set to 4: 1 in 8VSB, 16VSB mode, respectively. 6. The method of claim 5, wherein the parameter is used in the G-PSEUDO algorithm. The channel equalization method, characterized in that each is set at a ratio of 1: 2 in 15VSB mode.