KR100556385B1 - Apparatus of digital broadcast receiving set and Method of digital broadcast receiving - Google Patents

Abstract

The present invention relates to a digital broadcast receiving system including a noise canceller using an MLSE signal. According to the present invention, a demodulator for demodulating a received RF signal into a baseband signal through carrier / timing recovery, etc., and the received signal is a transmitter. A frequency domain equalizer for compensating for various fadings occurring during the arrival to the receiver, a noise canceller for removing amplified noise during the equalization process, an output of the noise canceller, and generating a reliable determination signal MLSE, And a MLSE unit for feeding back the MLSE signal to the noise canceller, and generating an accurate reference noise signal using the MLSE signal even for a signal passing through a channel having a very large fading. When receiving digital broadcasts with noise canceller with improved performance compared to It has the effect of making a stem.

Noise Canceller, MLSE, LSM, Equalizer, Viterbi Decoder

Description

Apparatus of digital broadcast receiving set and method of digital broadcast receiving}

1 is a block diagram illustrating a receiver system using a general frequency domain equalizer.

2 is a diagram illustrating a frequency spectrum of a fading mixed channel and a received signal in a typical receiver.

3 shows spectral changes in signal and noise when a fading component is compensated using a general frequency domain equalizer.

4 is a block diagram illustrating a receiving system including a noise canceller according to the prior art.

5 is a detailed view of the noise canceller of the reception system according to FIG.

6 illustrates a receiving system including a noise canceller according to the present invention.

7 is a view showing the structure of a noise canceller using the MLSE output according to the present invention

8 illustrates a frequency spectrum of a receiver using a noise canceller according to the present invention.

-Explanation of symbols for the main parts of the drawing-

110: demodulator 120: frequency domain equalizer

130: noise canceller 131: first subtractor

133: second subtractor 135: filter unit

135: third subtractor 140: MLSE part

The present invention relates to a digital broadcast reception system, and more particularly, to a digital broadcast reception system including a noise canceller using an MLSE output.

When the ATSC (Advanced Television Systems Committee) 8VSB transmission system, which is currently used for the digital transmission system and the US digital TV transmission system, and the transmission system using PAM or QAM, are transmitted by air or wire, the digital TV receiver has many The signals reflected by the reflector are received in combination with the original signal transmitted. At this time, since the reflected components distort the original signal, the original signal cannot be obtained only by the received signal. Here, the components that distort the original signal include ghost or fading.

That is, the physical channel of the digital transmission system including the ATSC 8VSB transmission system is free space. This channel can be thought of as an ideal channel when there is no vacuum in the medium, but the actual transmission signal is received through multiple paths due to the deflection of paths propagated by weather changes such as air, clouds, atmospheric layers, or the effects of the sun. do.

In addition, a fixed multi-path that is reflected, transmitted and propagated by a terrain, a mountain, a cliff, a land, and the like, and a time-varying multi-path generated by an airplane, a car, or the like, may occur.

Such multipath propagation causes the same signal to be transmitted through multiple paths having different propagation times, thereby causing interference between symbols, which is the biggest factor that degrades the performance of a high-speed digital transmission system.

As described above, the distortion of the signal due to the multipath propagation causes a bit detection error in the digital TV receiver, so that the entire screen may be impossible to recover or an entirely different image may appear.

In order to overcome this phenomenon, a digital TV receiver uses an equalizer for compensating for a component (that is, ghost or fading) that distorts a transmission signal between the transmitter and the receiver.

The equalizer may be divided into a time-domain equalizer and a frequency-domain equalizer, all of which remove components that distort the original signal.

1 is a block diagram illustrating a receiver system using a general frequency domain equalizer.

As shown in FIG. 1, a demodulator 1 for demodulating a received RF signal into a baseband signal through carrier / timing recovery, and the received signal is transmitted from a transmitter to a receiver. It consists of a frequency domain equalizer 2 which compensates for each kind of fading occurring during its arrival.

The frequency domain equalizer 2 estimates an impulse response of a channel in a time domain signal, and then converts the estimated channel impulse response into a frequency domain signal to convert the frequency domain signal distorted by the channel into an ideal frequency domain signal. It serves to correct the distortion by adjusting the coefficients of the frequency domain equalizer 2.

Therefore, after the received signal in the time domain is converted into a signal in the frequency domain, the fading component is compensated by the frequency domain equalizer 2, and the compensated frequency domain signal is converted into a time domain signal again. .

Figure 2 shows the frequency spectrum of a faded mixed channel and a received signal in a typical receiver, where (a) is a channel impulse response with three fading signals delayed with the main signal, and (b) is an ideal transmission from the transmitter. It shows the frequency spectrum of the signal and the frequency spectrum in the presence of fading, plus the white noise AWGN.

As shown in FIG. 2, when a frequency impulse response of a channel is converted into a frequency domain signal, an ideal channel having a magnitude 1 is flat while a portion having a very low magnitude (spectrum null) occurs.

3 is a diagram illustrating spectral changes in signal and noise when a fading component is compensated using a frequency domain equalizer.

Figure 3 (a) is the spectrum of the signal and noise before equalization, it can be seen that the AWGN is added to the entire band with a constant size, the fading received signal can be seen that the size of the spectrum is very small at a specific frequency have.

3 (b) shows a frequency spectrum of a signal that has undergone frequency domain equalization, and the signal spectrum coincides with an ideal signal, but the noise of a portion where the signal spectrum is small before compensation is increased.

That is, fading is compensated by converting a signal that has undergone frequency domain equalization back to a time domain signal, but the noise of the entire signal increases due to the noise increase in a specific frequency section, thereby degrading performance.

For this reason, a noise canceller has been conventionally used to remove such noise.

The receiving system constructed using the noise canceller is shown in FIG.

4 is a block diagram illustrating a receiving system including a noise canceller according to the prior art.

4 is the form which added the noise canceller 3 to the structure of FIG. That is, in order to remove the noise generated through the frequency domain equalizer 2, the noise canceller 3 is provided at the rear end of the frequency domain equalizer 2.

The configuration of the noise canceller 3 is shown in FIG.

5 is a diagram illustrating in detail the noise canceller of the reception system according to FIG. 4.

As shown in FIG. 5, the input value is a signal output converted from the frequency domain equalizer 2 to the time domain after frequency domain equalization. The input value produces a reference error that must be compensated for by the difference from the signal determined by the VSB slicer 3e in the first subtractor 3a through the VSB slicer 3e. The reference noise is used as the input of the filter 3d.

In addition, the second subtractor 3b obtains a difference between the reference noise and the estimated noise and uses it as an error. The error is used as the error input of the filter 3d, and the third subtractor 3c increases in the frequency domain equalizer 2 by subtracting the noise estimated by the filter 3d from the noise canceler 3 input signal. Remove the noise.

In addition, the filter 3d is a filter using a coefficient update method of a Least Mean Square (LMS) method in the form of a tapped delay line.

The coefficient update scheme is as follows.

c (k, n + 1) = c (k, n) + u * e (n) * x (k, n)

Here, n represents an integer from 0 to n, and k represents a k-th tap coefficient. U is an integer that determines the magnitude of the value to be updated, e (n) is the error value to be input, and x (k, n) is the input value of the k-th tap at n time.

On the other hand, as mentioned above, the slicer 3e uses various types of algorithms that can determine the VSB signal as a part for determining the VSB signal.

By using the noise canceller having such a configuration, the amplified noise in the frequency domain equalizer 2 is removed.

However, when a signal having a very large fading is received, the noise amplification is also very large so that the reference noise signal used in the noise canceller 3 becomes inaccurate, which causes the noise canceller to not operate efficiently. There was.

Accordingly, an object of the present invention is to solve such a problem of the prior art, by using the MLSE output that generates a reliable decision signal from the output signal of the noise canceller, the noise that operates efficiently in a very large fading environment Disclosed is a digital broadcast reception system having a canceller.

A digital broadcast reception system including a noise canceller according to the present invention for achieving the above object is a demodulator for demodulating the received RF signal into a baseband signal through a carrier / timing recovery, etc., the received signal is a receiver at the transmitter A frequency domain equalizer for compensating for various fadings occurring during the process, a noise canceller for removing amplified noise during the equalization process, an output of the noise canceller, and generating an MLSE that is a reliable determination signal. And a MLSE unit for feeding back a signal to the noise canceller.

The noise canceller may include: a first subtractor configured to delay a signal including noise amplified by the equalizer and a signal for determining an output signal of the noise canceller, and obtain a difference to generate a first reference noise signal; A second subtractor for generating an error signal through a difference between a reference noise signal and a signal delayed by the noise signal predicted through the noise canceller internal filter unit, and using the first reference noise signal and the error signal in a coefficient update scheme. A first LMS filter for predicting noise, a plurality of filter subtractors each generating a plurality of second reference noise signals by delaying and subtracting each of the MLSE signal generated through the MLSE unit and the signal including the amplified noise; A plurality of second LMS filters for predicting noise by a coefficient update method using the second reference noise signal and the error signal; An adder that adds predicted noise through a first LMS filter and a plurality of second LMS filters to produce one predicted noise, and a third subtracts the predicted noise added through the adder from a signal containing the noise amplified by the equalizer It is characterized by including a subtractor.

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

6 is a diagram illustrating a receiving system including a noise canceller according to the present invention.

As shown in FIG. 6, a reception system including a noise canceller according to the present invention includes a demodulator for demodulating a received RF signal into a baseband signal through carrier / timing recovery or the like ( 110, a frequency domain equalizer 120 for compensating for various fadings occurring while the received signal reaches the receiver, and a noise canceller 130 for removing amplified noise during the equalization process; The MLSE unit 140 receives the output of the noise canceller 130, generates a maximum likelihood sequence estimation (MLSE) that is a reliable determination signal, and feeds the MLSE signal back to the noise canceller 130. It consists of. The MLSE signal is generated using a Viterbi decoder in the case of VSB modulation.

According to the present invention, by using the output signal of the MLSE unit 140 to generate a reference noise signal of the noise canceller 130 to increase the noise removal performance, to look at the relationship in more detail Referring to Figure 7 attached to the structure of the noise canceller as follows.

7 is a diagram illustrating the structure of a noise canceller using the MLSE output according to the present invention.

As shown in FIG. 7, the structure of the noise canceller 130 according to the present invention is the same as that of FIG. 5 except for the filter unit 135 of the drawing. In other words, the structure of the filter part is changed by using the MLSE output.

Looking at the operation of the noise canceller according to the present invention, configured as shown in Figure 7 as follows.

The signal input to the noise canceller 130 is a signal having amplified noise as an output signal of the frequency domain equalizer 120.

The input signal having the amplified noise is input to the first subtractor 131 and the filter unit 135 of the noise canceller 130.

The first subtractor 131 obtains a difference between the input signal and the delayed signal from the sliced data determined by a slicer (not shown), and generates a reference error to be compensated for.

The reference noise is used as the (12 + m) tap length LMS filter 135a input of the filter unit 135.

In addition, by delaying the reference noise and the noise predicted from the filter unit 135 and subtracting it from the second subtractor 133, the difference is used as an error, and the error is (12 + mtap) length LMS filter ( 135a), as well as a number of 12 tap length LMS filters (135b, ..., 135c) is used as the error input.

That is, the LMS filter predicts an error by using a tapped-delay-line least mean square (LMS) coefficient update method using the reference error and an error.

The coefficient update scheme is as follows.

c (k, n + 1) = c (k, n) + u * e (n) * x (k, n)

Here, n represents an integer from 0 to n, and k represents a k-th tap coefficient. U is an integer for determining the magnitude of the value to be updated, e (n) is an error value to be input, and x (k, n) is an input value of the k-th tap at n time.

Instead of the LMS filter, another type of filter capable of noise prediction may be used.

On the other hand, the MLSE determination signal fed back from the MLSE unit 140 is output for every 12 symbol time delays and input as the fb_mlse signal of FIG. 7.

At this time, the input signal of the noise canceller 130 input to the filter unit 135 is delayed by (12 + m) D in the (12 + m) D block 135d, where (12 + m) D is the first input. The signal delay, m is the time from the output of the noise canceller to obtain the MLSE output signal.

The input signal delayed by (12 + m) D and the fb_mlse determination signal fed back from the MLSE unit 140 are delayed, respectively, and the difference is obtained through the filter first subtractor 135e, thereby having an LMS filter 135a having a length of 12 taps. Becomes a reference noise signal.

The LMS filter 135a uses the reference noise signal and the previously obtained error signal as input values to predict noise by an LMS coefficient update method.

In addition, the signal delayed by (12 + m) D is again delayed by 12 symbols in the 12D block 135f, and the LMS having each 12tap length separated by a difference from the signal fb_mlse fed back from the MLSE unit 140. The reference noise signal input value of the filters ... The noise is predicted in the same manner based on the reference noise signal and the previously obtained error signal.

This process is repeated by the number of LMS filters having a length of 12 taps, and as the number of filters increases, the noise reduction performance increases.

The noise predicted at intervals of 12 taps through the LMS filters 135a,..., And 135c is summed through the adder 135h to become predicted noise.

The predicted noise is used to find the difference with the input signal of the noise canceller 130 through the third subtractor 137. That is, by subtracting the predicted noise from the input signal of the noise canceller 130 having the amplified noise to output a signal from which the noise is removed.

As such, as the reference noise signal using the feedback of the reliable MLSE output signal is input, improved performance is obtained compared to the conventional noise canceller.

Table 1 below shows the performance improvement of the noise canceller using the MLSE output proposed in the present invention.

Conventional noise canceller Noise canceller of the present invention Receivable range 3us 5us

The results in the table above measure the acceptable range for a 0 ghost, single ghost in a VSB receiving system. In the conventional noise canceller, the maximum delay time of the ghost is 3us, and the noise canceller proposed in the present invention has a maximum delay time of 5us, so that the noise canceller of the present invention shows improved performance.

8 is a diagram illustrating a frequency spectrum of a receiver using a noise canceller according to the present invention.

FIG. 8 (a) is a diagram illustrating a spectrum after equalization in a signal that has undergone conventional frequency domain equalization, and shows that there is an increase in noise. FIG. 8 (b) shows a case where the noise canceller according to the present invention is performed. In the figure, it can be seen that the overall noise level is reduced by removing the noise increase component.

As described above, an apparatus and method for receiving digital broadcast including a noise canceler according to the present invention have the following effects.

First, an accurate reference noise signal is generated using an MLSE signal even for a signal passing through a channel having a very large fading, thereby producing a noise canceller having improved performance compared to a conventional noise canceller.

Second, by using the noise canceller having the improved performance in the reception system, there is an effect of improving the performance of the entire digital broadcast reception system.                     

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 examples, but should be defined by the claims.

Claims (5)

A demodulator for demodulating the received RF signal into a baseband signal through a carrier / timing recovery; A frequency domain equalizer for compensating for various fadings occurring while the received signal arrives from the transmitter to the receiver; A first subtractor for delaying a signal including the noise amplified by the equalizer and a signal determining the output signal of the noise canceller, and obtaining a difference to generate a first reference noise signal; A second subtractor for generating an error signal through a difference between the first reference noise signal and a signal delaying the predicted noise signal; A first LMS filter for predicting noise by a coefficient update method using the first reference noise signal and the error signal; A plurality of filter subtractors each generating a plurality of second reference noise signals by delaying and subtracting the maximum likelihood sequence estimation (MLSE) signal of the signal including the amplified noise of the equalizer and the signal including the amplified noise, respectively; , A plurality of second LMS filters for predicting noise by a coefficient update method using the second reference noise signal and the error signal; An adder for adding a predicted noise through the first LMS filter and a plurality of second LMS filters to produce a single predictive noise, And a third subtractor which subtracts the prediction noise added by the adder from the signal including the noise amplified by the equalizer. The method of claim 1, And a maximum likelihood sequence estimation (MLSE) signal of the signal including the amplified noise of the equalizer is generated by a Viterbi decoding method. delete Demodulating the received RF signal into a baseband signal through carrier / timing recovery, etc .; An equalization step of compensating for various fadings occurring while the received signal arrives from the transmitter to the receiver; Generating a first reference noise signal by delaying a signal including the amplified noise and a signal from which the noise has been removed during the equalization process and obtaining a difference therebetween; Generating an error signal through a difference between the first reference noise signal and a signal delaying the predicted noise signal; Predicting noise by an LMS coefficient update method using the first reference noise signal and an error signal; Generating a plurality of second reference noise signals by delaying and subtracting a maximum likelihood sequence estimation (MLSE) signal of the signal including the noise and the signal including the amplified noise, respectively; Predicting noise using an LMS coefficient update method using the second reference noise signal and the error signal; Adding all of the predicted noises to produce a single predictive noise and outputting the error signal for noise generation and noise cancellation; And removing the noise by subtracting the prediction noise from the signal including the amplified noise in the equalization process. delete

Images