CN1115875C - Digital TV signal receiver, method for processing AM signal with vestigial sideband and determining whether co-channel interference suppression filter is used - Google Patents

Abstract

An NTSC co-channel interference detector detects interfering NTSC signals present in a received Q-channel signal in orthogonal relationship to said received I-channel signal instead of detecting interfering NTSC signals present in a received I-channel signal. By determining whether a significant amount of NTSC co-channel interference is present in the received Q channel signal, it is inferred whether a significant amount of NTSC co-channel interference is present in the received I channel signal, and therefore, a large number of errors generated during trellis decoding of the equalized received I channel signal can be corrected during Reed-Solomon decoding following the trellis decoding. Accurate determination of co-channel NTSC interference levels is simplified because quadrature phase synchronous detection of the pilot carrier in the VSB AM digital television signal produces substantially no dc bias.

Description

Digital TV signal receiver, method for processing vestigial sideband AM signal and determining whether co-channel interference suppression filtering is used

The present invention relates to a digital television system, in particular to a circuit used by a digital television receiver to determine whether there is co-channel interference in an NTSC analog television signal.

Background of the invention

On September 16, 1995, the Advanced Television Standards Committee (ATSC) announced the digital TV standard to transmit 6MHz bandwidth TV channel digital TV (DTV) signals, such as the National Television Standards Committee (NTSC) analog TV signal used in air broadcasting in the United States. The vestigial sideband (VSB) signal is the same. The spectrum of the VSB DTV signal so designed may interleave with the spectrum of the co-channel interfering NTSC analog TV signal. This is accomplished by setting the pilot carrier frequency and primary AM sideband frequency of the DTV signal to an odd multiple of the 1/4 line rate of the NTSC analog TV signal, the "odd bits" being at 1/4 of the NTSC analog TV signal Between the "even multiples" of the 4-line scan line rate, at which most of the energy of the luminance and chrominance components of the co-channel interfering NTSC analog TV signal is attenuated. The video carrier frequency of the NTSC analog TV signal deviates 1.25MHz from the lower limit frequency of the TV channel. The frequency offset between the carrier frequency of the DTV signal and the video carrier frequency is 59.75 times of the horizontal scan line rate of the NTSC analog TV signal, so that the carrier frequency of the DTV signal deviates from the lower limit frequency of the TV channel by about 309877.6kHz. Correspondingly, the carrier frequency of the DTV signal deviates from the middle frequency of the TV channel by about 2690122.4 Hz.

The exact symbol rate specified by the digital television standard is 684/286 times the frequency offset (4.5MHz) between the sound carrier frequency and the video carrier frequency in the NTSC analog TV signal. Among them, 684 is the number of symbols in each line scanning line of each NTSC analog TV signal, and 286 is the multiplication coefficient of the line rate of the line scanning line of the NTSC analog TV signal. The function of this coefficient is to make the sound carrier frequency of the NTSC analog TV signal deviate from its The video carrier frequency is 4.5MHz. The symbol rate is 10.762238 megasymbols per second, which can be contained by the VSB signal which is 5.38119 MHz away from the DTV signal carrier. That is to say, the VSB signal can be limited to a frequency band whose boundary frequency is 5.690997 MHz away from the lower limit frequency of the medium video channel.

In the United States, a digital HDTV signal, the ATSC standard for which is used for terrestrial broadcasting, is capable of transmitting two high definition television (HDTV) formats with an aspect ratio of 16:9. One HDTV display format uses 1920 samples per scan line, 30 Hz frames with 2:1 field spacing, and 1080 active horizontal scan lines per frame. Another HDTV display format uses 1280 brightness sampling points per scan line, 60Hz frame, and 720 progressive scan lines per frame. In addition to HDTV display formats, the ATSC standard also includes the transmission of DTV display formats, such as the parallel transmission of four television signals with normal definition compared to NTSC analog television signals.

The DTV transmitted by vestigial sideband (VSB) amplitude modulation (AM) during terrestrial broadcasting in the United States includes a series of time-continuous data fields, and each time-continuous data field contains 313 time-continuous data segments. The data fields can be regarded as consecutively numbered modulo 2, and each odd-numbered data field and the subsequent even-numbered data field form a data frame. The frame rate is 20.66 frames per second. The duration of each data segment is 77.3 microseconds, so when the symbol rate is 10.76MHz, each data segment contains 832 symbols. Each data segment begins with a row sync block containing four symbols with values +S, -S, -S, and +S. The level of the +S value is lower than the maximum positive data amplitude, and the level of the -S value is higher than the maximum negative data amplitude. The first line of each data field includes a field synchronization code group for encoding the training signal, which is used for channel equalization and multipath suppression. The initial part of the training signal is a pseudo-noise sequence (or "PN sequence") containing 511 samples, and the following part is three PN sequences each containing 63 samples. The middle part of the 63-sample PN sequence in the field synchronization code is transmitted according to the first logical specification of the first row of each odd-numbered data field and the second logical specification of the first row of each even-numbered data field. The first logical specification and The second logical specifications are complementary to each other.

The data in the data row is trellis encoded with 12 interleaved trellis codes, each 2/3 rate trellis code has one uncoded bit. Interleaved trellis codes are subject to Reed-Solomon forward error correction coding, which corrects bursts caused by noise sources such as nearby unshielded automobile ignition systems Bit errors (burst errors). For air transmission, the Reed-Solomon encoding result is transmitted in an 8-level 3-bit/symbol one-dimensional constellation (one-dimensional-constellation) symbol encoding mode, and this transmission is performed without symbol precoding Yes, the symbol precoding process is separate from the trellis coding process. For cable television transmission, the Reed-Solomon encoding result is transmitted in a 16-level (4 bits/symbol) one-dimensional constellation symbol encoding mode, and this transmission is performed without symbol precoding. The VSB signal suppresses the inherent carrier whose amplitude varies with the modulation percentage.

The natural carrier can be replaced by a fixed-amplitude pilot carrier whose amplitude corresponds to the specified modulation percentage. The fixed-amplitude pilot carrier is generated by injecting a DC bias component into the modulation voltage applied to a balanced modulator that produces AM sidebands that are filtered by a filter to produce the VSB signal . If the 8 levels encoded by the 4-bit symbol in the carrier modulation signal have normalized values -7, -5, -3, -1, +1, +3, +5 and +7, then the normalization of the pilot carrier signal The normalized value is 1.25. +S has a normalized value of +5, and -S has a normalized value of -5.

In the early days of DTV technology, it was envisaged that when a DTV broadcaster was used with a comb filter in each DTV signal receiver (which is located before the data slicer in the symbol demultiplexer circuit as a postcoder), it must It is up to the DTV broadcaster to decide whether to use the symbol precoder at the transmitter, which follows the symbol generation circuit and provides matched symbol filtering. This decision can only rely on whether interference from co-channel NTSC broadcasting stations is predicted. Symbol precoding operations shall not be possible in data line sync block or between data lines that transmit data field sync data.

The farther away from the NTSC broadcast station, the smaller the co-channel interference is, and the co-channel interference is more likely to occur when certain ionospheric conditions are met. Co-channel interference is likely to occur during the summer of years of high solar activity. Of course, there would be no such interference if there were no co-channel NTSC broadcast stations. If NTSC interference is likely to occur within the broadcast coverage area, it is assumed that HDTV broadcasters will use symbol precoders to more easily separate HDVT signals from NTSC interference signals; thus, comb filters will be used as The symbolic post-encoder to perform matched filtering. If there is no possibility or no NTSC interference at all, in order to avoid misjudgment of the trellis coder symbol values caused by white noise, it is assumed that the DTV broadcaster will stop using the symbol precoder; thus, each A DTV signal receiver cannot use a symbol postcoder.

U.S. Patent No. 5,260,793, issued Nov. 9, 1993 to R.W. Citta et al., entitled "RECEIVER POSTCODER SELECTION CIRCUIT (Receiver Postcoder Selection Circuit)" employs a postcoder comb filter to suppress the NTSC interference that occurs simultaneously with active or in-phase baseband components (I channel) in the composite output signal of a demodulator in a (HDTV) receiver. By detecting whether there is an NTSC interference signal in the I channel component output by the demodulator, the control signal automatically starts or stops the work of the comb filter for suppressing NTSC co-channel interference. In each data field synchronization interval, the input signal and the output signal of the comb-shaped NTSC rejection filter in the HDTV signal receiver are respectively compared with the correlation signal determined in advance and extracted from the memory of the HDTV signal receiver. If the energy of the minimum comparison result compared with the input signal is less than the energy of the minimum comparison result compared with the output signal of the NTSC rejection filter, it means that the main cause of the difference from the expected received signal is random noise rather than NTSC co-channel interference signal. As far as certain digital television receivers are concerned, the received signal is preferably pre-coded and no post-coding is used in the system, so it is assumed that the broadcaster does not use pre-coding. If the energy of the minimum comparison result compared with the input signal is greater than the energy of the minimum comparison result compared with the output signal of the NTSC rejection filter, it means that the main cause of the difference from the expected received signal is the NTSC co-channel interference signal rather than random noise. As far as certain digital television receivers are concerned, the received signal is preferably pre-coded and post-coded in the system, so it is assumed that pre-coding has been used by the broadcaster.

U.S. Patent No. 5,546,132, issued August 13, 1996 to K.S. Kim et al., entitled "NTSC INTERFERENCEDETECTOR (NTSC Interference Detector)" describes the use of post-encoded comb filtering to suppress the NTSC extraction comb filter in response to the I channel co-channel NTSC interference signal detected at the same time. US Patent No. 5,546,132 does not specifically address reactive or quadrature baseband components (Q channels) in the composite output signal of a demodulator used in a digital HDTV signal receiver. A digital HDTV signal receiver that synchronizes a VSB AM signal to baseband typically employs a demodulator that includes an in-phase sync detector and a quadrature-phase sync detector, where the in-phase sync detector is used to provide the received The I channel signal is used for trellis decoding (after postcoding, if precoding is used in the transmitter), and the quadrature phase sync detector is used to provide the received Q channel signal. The received Q-channel signal is low-pass filtered to generate an automatic frequency and phase control (AFPC) signal to allow a local oscillator to provide a synchronized carrier. No. 5479449 entitled "DIGITAL VSB DETECTOR WITH BANDPASS PHASE TRACKER, AS FOR INCLUSION IN AN HDTV RECEIVER" awarded to C.B. Patel and A.L.R. Limberg on December 26, 1996 The specification and drawings of U.S. Patent No. , assigned to Samsung Electronics Corporation, are hereby incorporated by reference. The reader should pay particular attention to components 22-27 shown in Figure 1 of the drawings of US Pat. No. 5,479,449 and the description of these components in the patent specification. These components are used in the described HDTV signal receiver for composite demodulation of the VSB AM final intermediate frequency (IF) signal. US Patent No. 5479449 describes the complex demodulation of the VSB AM final I-F signal under digital conditions, but the complex demodulation of VSB AM final I-F signals in other digital TV receivers is performed under analog conditions.

In U.S. Patent No. 5,260,793 and U.S. Patent No. 5,546,132, post-coding is allowed during the absence of co-channel NTSC interfering signals, otherwise post-coding is not allowed. This selective activation control signal is based on the received I channel signal obtained. It is difficult to determine the level of the same-channel NTSC interference signal due to the DC bias associated with the same-channel NTSC interference signal, which is caused by the in-phase synchronous detection of the pilot carrier signal in the VSB AM DTV signal. This is a significant problem in DTV signal receivers where the automatic gain control signal does not tightly adjust the amplitude of the received I-channel signal recovered by in-phase sync detection.

The video carrier frequency of the NTSC signal is 1.25 MHz from the edge of the 6 MHz wide broadcast channel, while the carrier frequency of the DTV signal used for terrestrial broadcasting is 310 kHz from the edge of the 6 MHz wide broadcast signal. For vestigial sideband amplitude modulated (VSB AM) carriers carrying digital information, co-channel NTSC signals do not exhibit symmetrical AM sidebands. Thus, the 940 kHz NTSC video carrier artifact is removed from the DTV signal carrier, but its sideband artifacts are not completely removed from the DTV signal due to synchronization with baseband. Of course, the artifacts of the NTSC audio carrier and its sidebands have not completely disappeared from the DTV signal, and the NTSC audio carrier at 5.44 MHz has disappeared from the DTV signal carrier.

The digital TV standard ATSC announced on September 16, 1995 did not consider precoding all data in the DTV transmitter to compensate for the aftereffects that are prone to occur when comb filtering is used in the DTV signal receiver to suppress NTSC co-channel interference. encoding, but only pre-encodes the original symbols in trellis decoding. This process by itself does not help DTV signal receivers use comb filters to suppress NTSC co-channel interference before the data clipping process is performed. DTV digital receivers that are unable to suppress NTSC co-channel interference artifacts before the data clipping process is performed in cases of strong NTSC co-channel interference that may be caused by the DTV digital signal receiver being far from the DTV digital transmitter or by analog TV transmitters nearby The next will not be well received. In a baseband synchronous DTV signal, the artifact frequency of the co-channel interfering video carrier in the NTSC color TV signal is 59.75f _H , where f _H is the horizontal scanning frequency of the NTSC color TV signal. The artifact frequency is _287.25fH for the color subcarrier and 345.75fH for the unmodulated NTSC audio carrier. The inventors have pointed out that the comb filtering process is not entirely suitable for eliminating carrier artifacts in FM NTSC audio, especially in the case of FM with a relatively large carrier frequency drift, because the FM obtained at any fixed-delayed sampling instant It is possible that the correlation (or anti-correlation) of the carrier samples is not particularly good. The inventors have proposed that the filtering used to amplify the entire IF bandwidth must facilitate rejection of the FM audio carrier in any co-channel NTSC analog TV signal. The comb filtering process is well suited for separating baseband DTV signals from artifacts of the NTSC video carrier, low video frequencies, and chrominance signal frequencies near the color carrier. This is because some of these artifacts have better correlation between samples at certain delay intervals and better anti-correlation between samples at certain other delay intervals.

On November 12th, 1996, the inventor proposed in the United States entitled "DTV RECEIVERWITH FILTER IN I-F CIRCUITRY TO SUPPRESS FM SOUND CARRIER OFNTSC CO-CHANNEL INTERFERING SIGNAL (with I-F for suppressing FM acoustic carrier and NTSC co-channel interference signal DTV receiver of the filter in the circuit) "No. 08/746520 patent application, the inventor proposes, when the NTSC co-channel interference is large enough to affect the data clipping conversely, just through the comb in the DTV signal receiver The shape filter performs data clipping first, thereby suppressing NTSC co-channel interference. The inventors have also shown how to compensate for the effect of this comb filtering on selectively performed symbol encoding during symbol decoding. In addition, it can also determine the moment when the NTSC co-channel interference is greater than a specified value, which is the minimum acceptable value, so that the determination result can be used to control the selective use of the comb filter for NTSC co-channel interference. Interference is suppressed.

Whenever NTSC co-channel interference is present in the active or in-phase baseband component (I channel) of the composite output signal of a demodulator used in a DTV signal receiver, it is also present in its reactive or quadrature-phase baseband component (Q channel) co-channel interference. Accordingly, an NTSC jammer detector can be arranged such that its NTSC extraction filter is responsive to the received Q-channel signal rather than the received I-channel signal. By determining whether there is a large value of NTSC co-channel interference in the received Q-channel signal, it can be inferred whether there is a large value of NTSC co-channel interference in the received I-channel signal, so that it can be decoded by trellis The Reed-Solomon decoder behind the decoder corrects a large number of bit errors in the trellis decoding process of the equalized received I-channel signal. Accurate determination of co-channel NTSC interference levels is simplified because synchronous detection of the quadrature phase of the VSBAM DTV signal pilot carrier introduces substantially no DC bias.

Summary of the invention

A method for processing a vestigial sideband AM digital television signal in a digital television signal receiver according to an aspect of the present invention comprises the steps of: complex decomposing the vestigial sideband AM digital television signal susceptible to co-channel NTSC interference tune, so as to separate the received I-channel baseband signal and separate the received Q-channel signal which is in an orthogonal relationship with the received I-channel baseband signal; then, by determining the accompanying co-channel Whether the NTSC interference artifacts exceed the specified value is used to estimate whether the accompanying co-channel NTSC interference artifacts in the received I-channel baseband signal reach an effective level.

A method according to the present invention for determining whether comb filtering is used in a digital television receiver prior to trellis decoding to suppress co-channel NTSC interference comprises the steps of: complex demodulating the digital television signal in order to separate the received I channel baseband signal and separate the received Q channel signal which is in quadrature relationship with the received baseband signal; determine whether there is a co-channel NTSC interference artifact with a significant level in the received Q channel baseband signal; if determined There is no effective level co-channel NTSC interference artifact in the received Q-channel baseband signal, so that the received I-channel signal is directly decoded without comb filtering, thereby generating If it is determined that there is a significant level of co-channel NTSC interference artifacts in the received Q-channel baseband signal, the received I-channel baseband signal is comb-filtered to produce a comb-filtered I-channel baseband signal (in which co-channel NTSC interference has been suppressed), and the comb-filtered I-channel baseband signal is symbol-decoded, and then the symbol-decoded result of the comb-filtered I-channel baseband signal is post-coded , resulting in decoded symbols for trellis decoding.

According to a kind of digital television signal receiver of the present invention, comprise: amplifier circuit, be used for providing the amplified vestigial sideband AM digital television signal that is easy to carry co-channel interference analog television signal; The vestigial sideband amplitude modulated digital television signal responds to provide an I channel baseband signal containing any artifacts of the co-channel interfering analog television signal and a Q channel baseband signal containing any other artifacts of the co-channel interfering analog television signal; said I channel means for symbol decoding of baseband signals, comprising a first data slicer for symbol decoding said I-channel baseband signal for a first time, usually as long as said artifacts of any co-channel interfering analog television signal are below said The active level of the baseband signal of the I channel, then the bit error in the first symbol decoding result output by the first data limiter can be corrected; and the NTSC co-channel interference detector responding to the baseband signal of the Q channel , for detecting the occurrence of any of said other artifacts of the co-channel interfering analog television signal which are higher than the active level of said Q channel baseband signal, said active level of said Q channel baseband signal being equal to The active levels of the I-channel baseband signals are consistent.

An NTSC co-channel interference detector embodying aspects of the present invention detects interfering NTSC signals present in the Q channel, which is orthogonal to the I channel. Adaptive NTSC co-channel interference suppression circuitry further embodying aspects of the present invention utilizes these NTSC co-channel interference detectors to control whether comb filtering is used to suppress NTSC co-channel interference prior to data limiting in digital television receivers.

Brief description of the drawings

Figure 1 is a block diagram of a portion of a digital television receiver including a symbol decoder having an NTSC co-channel interference suppression circuit in accordance with the present invention based on NTSC co-channel interference responsive to a Q channel signal output from a DTV signal composite demodulator The detector responds to the signal to work;

Fig. 2 is the block diagram of the NTSC co-channel interference detector for responding to the Q channel signal that the DTV signal compound demodulator outputs according to the present invention;

FIG. 3 is an operational flowchart of the portion of the digital television receiver of FIG. 1 showing how the equalization operation is modified depending on whether comb filtering is used to suppress co-channel NTSC interference;

Fig. 4 is a detailed schematic block diagram of the digital circuit shown in Fig. 1 when the NTSC suppression comb filter adopts 12 symbol delays;

Fig. 6 is a detailed schematic block diagram of the part of the DTV signal receiver shown in Fig. 1 when the NTSC suppression comb filter adopts 6 symbol delays;

Figure 7 is a detailed schematic block diagram of a portion of the NTSC co-channel interference detector shown in Figure 2 when a 6-symbol delay is used;

Fig. 8 is the detailed block diagram of the part of DTV signal receiver shown in Fig. 1 when NTSC suppresses comb filter and adopts 2 video line delays;

Figure 9 is a detailed schematic block diagram of a portion of the NTSC co-channel interference detector shown in Figure 2 when a 2 video line delay is used;

Fig. 10 is a detailed schematic block diagram of the part of the DTV signal receiver shown in Fig. 1 when the NTSC suppression comb filter adopts 262 video line delays;

Figure 11 is a detailed schematic block diagram of a portion of the NTSC co-channel interference detector shown in Figure 2 when a 262 video line delay is employed;

Fig. 12 is a detailed schematic block diagram of the part of the DTV signal receiver shown in Fig. 1 when the NTSC suppression comb filter adopts 2 video frame delays;

Figure 13 is a detailed schematic block diagram of the NTSC co-channel interference detector portion shown in Figure 2 when using 2 video frame delays;

Figures 14 and 15 respectively represent the detailed schematic block diagrams of other types of NTSC co-channel interference detectors that can be used on the DTV signal receiver shown in Figure 1; and

Figure 16 is a schematic block diagram of a digital television receiver in accordance with the present invention in which a plurality of comb filters and their associated NTSC co-channel interference detectors are employed to selectively filter out NTSC co-channel interference detectors. Channel jamming artifacts.

Detailed description of the embodiment of the invention

It is understandable to those familiar with electronic circuit design that shimming delays must be inserted at various positions in the circuit shown in the accompanying drawings. The interstitial delay will not be involved in the following description unless there is a special requirement for the interstitial delay.

Figure 1 shows a digital television signal receiver for recovering error corrected data suitable for recording with a digital video cassette recorder (DVCR) or for MPEG-2 decoding and playback on a television. The DTV signal receiver among Fig. 1 receives TV broadcast signal from receiving antenna 8, but it also can receive signal from wired network. A television broadcast signal is used as an input signal to the "front end" circuit 10 . "Front-end" circuit 10 generally includes a radio frequency amplifier and a first detector for converting radio frequency television signals to intermediate frequency television signals. The intermediate frequency television signal serves as an input signal to an intermediate frequency (IF) amplifier circuit 12 which generates a vestigial sideband DTV signal. The DTV signal receiver is preferably a composite conversion type receiver with an IF amplifier circuit 12. The IF amplifier circuit 12 includes an IF amplifier, a second detector and another IF amplifier, wherein the IF amplifier is used to convert the first detector The resulting UHF DTV signal is amplified, a second detector is used to convert the amplified DTV signal into a very high frequency (VHF) signal, and another IF amplifier is used to amplify the converted DTV signal into a VHF signal. If demodulation is performed in the digital range, the IF amplifier circuit 12 will also include a third detector for converting the amplified DTV signal to a final intermediate frequency band closer to baseband.

UHF band IF amplifiers preferably use surface acoustic wave (SAW) filters to shape the channel selection response and reject adjacent channels. This SAW filter cuts off rapidly outside the 5.38MHz range from the suppressed carrier frequency of the VSB DTV signal and the pilot signal, which have the same frequency and fixed amplitude. This SAW filtering thus removes a substantial amount of any co-channel interfering FM sound carrier in the analog TV signal. Elimination of any co-channel interfering FM sound carrier in the analog TV signal in IF amplifier circuit 12 avoids artifacts of this carrier when detecting the final I-F signal to recover baseband symbols, and prevents these artifacts from interacting with those during symbol decoding. The data slices of the baseband symbols interfere with each other. Avoiding these artifacts that interfere with data clipping of baseband symbols during symbol decoding is better than comb filtering before data clipping, especially when different This is especially true for delays greater than a few symbol phase delays (epochs).

The final IF output signal output from the IF amplifier circuit 12 is provided to the composite demodulator 14, which demodulates the vestigial sideband AM DTV signal in the final intermediate frequency band to recover the active baseband signal and the reactive baseband Signal. Demodulation can be performed in the digital range after analog-to-digital conversion of the final intermediate frequency band in the several MHz range, as described for example in US Pat. No. 5,479,449. On the other hand, it is also possible to perform demodulation in the analog range. In this case, it is generally necessary to perform analog-to-digital conversion on the demodulation result for further processing. Composite demodulation is best accomplished with in-phase (I) synchronous demodulation and quadrature-phase (Q) synchronous demodulation. The digital result of the demodulation operation described above typically has 8-bit or better precision, which describes the 2N level symbols encoding N-bit data. Generally, 2N is equal to 8 when the DTV signal receiver in FIG. 1 receives over-the-air broadcast signals through the antenna 12, and 2N is equal to 16 when the DTV signal receiver in FIG. 1 receives cable TV signals. The present invention is related to receiving ground-to-air broadcast signals, so Fig. 1 does not show the symbol decoding part of the DTV signal receiver and the error correction decoding part for receiving cable TV transmission signals.

Symbol synchronization and equalization circuit 16 receives at least digitized active samples of the in-phase (I channel) baseband signal from complex demodulator 14; in the DTV signal receiver shown in FIG. 1 , circuit 16 also receives quadrature phase (Q channel) Digitized reactive samples of the baseband signal. Circuit 16 includes digital filters with adjustable weighting coefficients for compensating for ghosting and skew in the received signal. Symbol synchronization and equalization circuit 16 synchronizes or "de-rotates" symbols and equalizes amplitudes and eliminates ghosting. Symbol synchronization and equalization circuits are known from US Pat. No. 5,479,449, in which symbol synchronization is performed prior to amplitude equalization. In this design, demodulator 14 will provide an oversampled demodulator response signal containing active and reactive baseband signals to symbol synchronization and equalization circuit 16 . After symbol synchronization, in order to reduce the sampling rate of the amplitude equalization and ghost suppression digital filter, the oversampled data is de-sampled by 10, so as to extract the baseband I channel signal of the normal symbol rate. Symbol synchronization and equalization circuits in which amplitude equalization is performed prior to symbol synchronization, "derotation" or "phase tracking" are also well known to those skilled in the art of digital signal receiver design.

Each sample value of the output signal of circuit 16 is resolved into 10 or more bits, which are effectively a digital description of a single-level (one of 8 levels) analog symbol. The output signal of circuit 16 can be carefully controlled by any of several known gain control methods, so that the typical step levels of the symbols are known. A preferred method of gain control (which is preferred due to its particularly fast response) is to adjust the DC component of the active baseband signal provided by composite demodulator 14 to a normalized level of +1.25. U.S. Patent No. 5,479,449, especially to C.B. Patel et al., entitled "AUTOMATIC GAIN CONTROL OFRADIO RECEIVER FOR RECEIVING DIGITAL HIGH-DEFINITIONTELEVISION SIGNALS (Automatic Gain Control for Radio Receivers Receiving High-Definition Television Signals) issued June 3, 1997 )" US Pat. No. 5,573,454 describes this gain control method, which is hereby taken as a reference method.

The output signal of the circuit 16 is used as the input signal of the data synchronization detection circuit 18, and the data synchronization detection circuit 18 is used for recovering the data field synchronization information F and the data segment synchronization information S from the equalized baseband I channel signal. Alternatively, the input signal to the data synchronization detection circuit 18 may be obtained before equalization.

The normal symbol rate samples of the equalized I-channel signal in the output signal of circuit 16 are used as the input signal to NTSC rejection comb filter 20 . The comb filter 20 includes a first delay device 201 and a linear combiner 202, wherein the first delay device 201 is used to generate a pair of symbol streams with 2N levels of different delays, and the first linear combiner 202 is used to linearly combine different delayed symbol streams to generate a response signal of the comb filter 20 . As described in US Patent No. 5260793, the first delay device 201 can provide a delay equal to 12 2N-level symbol periods, and the first linear combiner 202 can be a subtractor. Each sample of the comb filter 20 output signal is resolved into 10 or more bits, which is effectively a digital representation of a single-level (4N = one of 15 levels) analog symbol.

Assume that the symbol synchronization and equalization circuit 16 is used to suppress a DC bias component in its input signal (represented in digital samples) with a normalized level of +1.25, due to the detection of the pilot carrier, thus compounding This DC bias component appears in the active baseband signal of demodulator 14 . Thus, each sample of the output signal of circuit 16, which is the input signal to comb filter 20, is actually at some normalized level (-7, -5, -3, -1, +1, + 3. Numerical description of the analog symbols of +5 and +7). These symbol levels are named as "odd" symbol levels, and after being detected by the odd level data slicer 22, they produce intermediate coincident decoding results of 000, 001, 010, 011, 100, 101, 110 and 111 respectively .

Each sample of the comb filter 20 output signal is actually at some normalized level (-14, -12, -10, -8, -6, -4, -2, 0, + 2, +4, +6, +8, +10, +12+ and 14) numerical descriptions of the analog symbols. These symbol levels are named "even" symbol levels, and they are detected by the even level data limiter 24 to generate 001, 010, 011, 100, 101, 110, 111, 000, 001, 010, 011 respectively , 100, 101, 100 and 111 such precoded symbol decoding results.

Data slicers 22 and 24 may be of the so-called "hard decision" type data slicers assumed so far in the description, or may be of the so-called "hard decision" type used when implementing a Viterbi (Viterbi) decoding scheme. Soft decision" type limiter. It is also feasible to replace the odd-level data slicer 22 and the even-level data slicer 24 with a single data slicer, as long as the multiplexer connection circuit is used to move its position in the circuit and provide a means for modifying its A clipping range bias would suffice, but these schemes are best avoided due to their operational complexity.

Assuming that the above-mentioned symbol synchronization and equalization circuit 16 is used to suppress the DC bias component in its input signal (expressed in digital samples), the normalized level of this DC bias component is +1.25, due to the detection of the pilot carrier, the composite This DC bias component appears in the active baseband signal of demodulator 14 . On the other hand, the symbol synchronization and equalization circuit 16 is used to maintain the DC bias component in its input signal, which slightly simplifies the design of the equalization filter in the circuit 16 . In this case, the data slicing levels of odd level data slicer 22 are shifted to take into account the accompanying DC bias component in the data steps of its input signal. Assuming that the first linear combiner 202 is a subtractor, no matter whether the circuit 16 is used for suppression or for holding, the DC bias component in its input signal is very important for the data clipping level of the even level data clipper 24. is irrelevant. However, when the differential delay provided by the first delay device 201 makes the first linear combiner 202 an adder, considering the accompanying DC bias component in its input signal data ladder, the even-level data limiter 24 The data slicing level should be offset.

Following the data slicers 22 and 24, a comb filter 26 is used to generate a post-coding filter response signal responsive to the response signal of the pre-coding filter in comb filter 20. The comb filter 26 includes 3 input multiplexers 261, the second linear combiner 262 and the second delay device 263, the delay of the second delay device 263 is the same as the first delay in the comb filter 20 The delay of device 201 is the same. If the first linear combiner 202 is a subtractor, then the second linear combiner 262 is a modulo-8 adder; if the first linear combiner 202 is an adder, then the second linear combiner 262 is a modulo-8 subtractor. The first linear combiner 202 and the second linear combiner 262 may constitute respective read only memories (ROMs) to increase the speed of the linear combination operation sufficiently to support the sampling speed involved. The output signal of the multiplexer 261 is the response signal of the post-coding comb filter 26, which is delayed by the second delay device 263. The second linear combiner 262 combines the decoding result of the precoded symbol output by the even-level data limiter 24 with the output signal of the second delay device 263 .

Multiplexer 261, in response to one of the first, second and third states of the multiplexer control signal provided by controller 28, has its output signal reproduce one of its three input signals. an input signal. When the data synchronization detection circuit 18 receives the data field synchronization information F and the data segment synchronization information S in the equalized baseband I channel signal, the first input port of the multiplexer 261 receives the standard symbol decoding from the internal memory of the controller 28 result. During this period, the controller 28 inputs the first state of the multiplexer control signal into the multiplexer 261, so that the multiplexer 261 uses the standard symbol decoding result provided by the internal memory of the controller 28 as the final Decoding result (output signal). The odd level data slicer 22 provides the intermediate symbol decoding result to the second input port of the multiplexer 261 . The second state of the multiplexer control signal causes the final decoded result of multiplexer 261 to reproduce the intermediate symbol decoded result. The second linear combiner 262 provides a third input port of the multiplexer 261 with the decoded result of the post-encoded symbols as its output signal. The third state of the multiplexer control signal causes the multiplexer 261 to reproduce, for example, the post-encoded symbol decoding results. During the recovery of the data field synchronization information F and the data segment synchronization information S by the data synchronization detection circuit 18, the standard symbol decoding result provided by the internal memory of the controller 28 reduces the running error in the post-coding decoding result output by the post-coding comb filter 26. code.

The output signal of multiplexer 261 within post-coding comb filter 26 includes the final symbol decoded results in parallel 3-bit form, which is assembled by data assembler 30 and input to data interleaver 32 . Data interleaver 32 interleaves the assembled data into parallel data streams, which are then provided to trellis decoder circuit 34 . Trellis decoder circuit 34 typically employs twelve trellis decoders. The trellis decoding result output by the trellis decoder circuit 34 is input into the data deinterleaver circuit 36 for inverse transformation. The byte analysis circuit 38 converts the output signal of the data interleaver 36 into a Reed-Solomon error correction coded byte, which is then input to the Reed-Solomon decoder circuit 40, and the Reed-Solomon circuit 40 performs Reed-Solomon decoding on it , thereby generating an error-correcting byte stream and providing the error-correcting byte stream to the data de-randomizer 42 . The data de-randomizer 42 provides the reproduced data to other parts of the receiver (not shown). Other parts of a complete DTV signal receiver will include packet classifiers, audio decoders, MPEG-2 decoders, etc. The rest of the DTV signal receiver in a digital video tape recorder/player will include circuitry for converting the data into a recording format.

NTSC co-channel interference detector 44 provides controller 28 with an indication of the strength of NTSC co-channel interference which indicates whether NTSC co-channel interference will produce bit errors which cannot be corrected by data slicing by data slicer 22 . If the detector 44 indicates that NTSC co-channel interference is not of such magnitude, then the controller 28 will send multiplex signals to the multiplexer during periods other than when the data sync detection circuit 18 performs recovery operations for the data field sync information F and the data segment sync information S. 261 provides the second state of the multiplexer control signal. This causes the multiplexer 261 to reproduce the intermediate symbol decoding result provided by the odd-level data slicer 22 as its output signal. If the decoder 44 indicates that the NTSC co-channel interference is strong enough to generate a bit error that cannot be corrected in the data slicing by the data slicer 22, the controller 28 will perform the data field sync information F and data segment sync at the data sync detection circuit 18 The period other than the recovery operation period of the information S provides the third state of the multiplexer control signal to the multiplexer 261, which makes the multiplexer 261 reproduce the second line combination provided by the second linear combiner 262 The decoded result of the post-coded symbol is used as its output signal.

The invention disclosed in this specification and its drawings is characterized by an NTSC co-channel interference detector 44 which is responsive to artifacts of NTSC co-channel interference present in the Q channel output signal of the composite demodulator 14 of the DTV signal. . Detector 44 may be used to detect artifacts of NTSC co-channel interference in the Q-channel output signal of composite demodulator 44 extracted prior to symbol synchronization and equalization circuit 16, but FIG. Detector 44 of artifacts in the Q channel output signal extracted from the response of circuit 16.

Figure 2 illustrates one form that NTSC co-channel interference detector 44 may take in one embodiment of the invention. The Q channel output signal extracted from the response of the symbol synchronization and equalization circuit 16 is directly added to the node 440, or is added to the word point 440 after being filtered by the bandwidth selection filter 441, and the bandwidth selection filter 441 supplies the node 440 Provides a response to those portions of the Q channel output signal that are more likely to contain NTSC channel interference artifacts. The signal at node 440 is applied as an input signal to third delay means 442 to undergo a third delay. The third linear combiner 443 linearly combines the signal at the node 440 with the signal delayed by the third delay device 442 to generate a comb filter response suppressing NTSC co-channel interference artifacts. A fourth linear combiner 444 linearly combines the signal at node 440 with the signal delayed by the third delay means 442 to produce a response containing selected NTSC co-channel interference artifacts. One of the third and fourth linear combiners is a digital adder and the other is a digital subtractor, the choice of which depends on the delay provided by the third delay means 442. The comb filter response amplitude of the third linear combiner 443 output is detected by the amplitude detector 445, and the comb filter response amplitude of the fourth linear combiner 444 output is detected by the amplitude detector 446, and the amplitude detector The amplitude detection results measured at 445 and 446 are compared by an amplitude comparator 447 . Amplitude detector 447 provides an output bit that indicates whether the response of magnitude detector 446 substantially exceeds the response of magnitude detector 445 . This output bit is used to select operation of the multiplexer 261 between the second and third states. For example, this output bit of magnitude comparator 447 may be one of two control bits provided by controller 28 to multiplexer 261 in post-coding comb filter 26 of FIG. The bit is used to indicate whether the signal provided by the controller 28 is present in the multiplexer 261 response.

For example, magnitude detectors 445 and 446 may be envelope detectors with time constants of several data sample periods, so that the data component differences in their input signals eventually tend to low values, assuming they are random. The random noise difference in the responses of linear combiners 443 and 444 also eventually tends to zero. Thus, when amplitude comparator 447, which compares the amplitude detection responses of amplitude detectors 445 and 446, indicates that the difference in those responses is greater than a specified value, this indicates that any co-channel interfering analog television signal artifacts exceed Q The active level of the channel's baseband signal. The active level of the Q channel baseband signal is equivalent to the active level of the I channel baseband signal. Symbol decoding errors caused by simple data clipping of the I-channel baseband signal can be corrected by Trellis and Reed-Solomon as long as any co-channel interfering analog TV signal artifacts are below the effective level of the I-channel baseband signal to be corrected.

When the magnitude of the comb filter response output by the fourth linear combiner 444 (which contains selected NTSC co-channel interference artifacts) is substantially greater than the comb filter response output by the third linear combiner 443 (which suppresses the When the magnitude of the NTSC co-channel interference artifact) can be assumed, the difference can be assumed to be due to the presence of NTSC co-channel interference in the signal at node 440. In this case, the output bit provided by the magnitude comparator 447 renders the multiplexer 261 inoperable in its third state and thus is eliminated from the time the final symbol decoding result appears on the multiplexer 261. The decoded result of the post-encoded symbols from the second linear combiner 262 is dropped.

Depending on the delay length of the third delay element 442 and the design of the amplitude detectors 445 and 446, the bandwidth selection filter 441 may not be required or even desired. In addition to being envelope detectors, magnitude detectors 445 and 446 can also detect the offset energy between their input signal and the symbol encoding level (derived from the pilot carrier strength); Bandwidth selection filter 441 . If the delay length of the third delay element 442 cannot completely eliminate the artifacts of the NTSC voice carrier, but can completely eliminate the artifacts of the NTSC audio carrier and color subcarrier, and if the amplitude detectors 445 and 446 are envelope detector, then the bandwidth selection filter 441 can take the form of a finite impulse response (FIR) digital low-pass filter 4410 with a cutoff frequency not greater than 5.4 MHz as shown in FIGS. 7 , 9 , 11 , 13 and 14 .

FIG. 3 is a flowchart showing the modification of the equalization operation of the DTV signal filter in FIG. 1 according to whether comb filtering is used to suppress co-channel NTSC interference. The inventor pointed out that when same-frequency NTSC interference occurs during baseband symbol encoding, errors will be brought to the calculation of equalization filter Kernel (kernel) coefficients, unless special measures are taken to invalidate these artifacts during calculation.

In the initial step S1, the composite demodulator 14 in the DTV signal receiver shown in Figure 1 performs composite demodulation to the digital television signal continuously, so that the I channel baseband signal received is separated and is orthogonal to the received I channel baseband signal The received Q channel baseband signal. In the decision step S2, the NTSC co-channel interference detector 44 in the DTV signal receiver shown in Fig. 1 also performs composite demodulation to the digital television signal continuously, and determines whether there is an effective amount of co-channel NTSC in the received Q channel baseband signal interference.

The rms value of the same-channel NTSC interference in the DTV signal receiver means that under normal noise reception conditions, trellis decoding will generate a large number of errors, which will greatly reduce the error correction capability of the two-dimensional Reed-Solomon decoding behind trellis decoding. , thus causing a large number of bit error levels in the final recovered data. The rms value of co-channel NTSC interference in a specially designed DTV signal receiver is easily determined by prototype experiments of the receiver.

If it is determined in the judgment step S2 that there is no co-channel NTSC interference with effective value in the received Q channel baseband signal, then step S3 and step S4 are executed to produce the symbol decoding result for step S5, wherein step S3 uses For adjusting the Kernel weight value of digital equalization filter, so that equalize the response to I channel baseband signal; Step S4 is used for carrying out symbol decoding to the equalization filter response that is given by step S3; Step S5 is used for marking symbol decoding result structure decoding, in order to correct the bit errors. Trellis decoding step S5 is followed by Reed-Solomon decoding steps S6 and S7. Step S6 is used to correct bit errors in the trellis decoding result, and step S7 is used to remove the format of the Reed-Solomon decoding result.

If it is determined in the judgment step S2 that there is effective amount of co-channel NTSC interference in the received Q channel baseband signal, then step S8 is performed with a suitable comb filter to perform comb filtering on the received I channel baseband signal, Thus, a comb-filtered I-channel baseband signal is generated. In step S9, by adjusting the Kernel weight of the digital equalization filter, the response of the cascaded digital equalization filter and the comb filter conforms to the standard response of the filter cascade. Execute step S10 to perform sign decoding on the filter cascade response. Then step S11 is executed to post-decode the symbol decoding response, so as to obtain the corrected symbol decoding result for trellis decoding step S5. Trellis decoding step S5 is followed by Reed-Solomon decoding steps S6 and S7. Step S6 is used to correct bit errors in the trellis decoding result, and step S7 is used to remove the format of the Reed-Solomon decoding result.

In step S3, the auxiliary method for adjusting the Kernel weight of the digital equalization filter so as to equalize the response of the digital equalization filter is similar to the method for adjusting the Kernel weight of the digital equalization filter in the prior art. The adjustment is carried out as follows: Discrete Fourier transform (DFT) is performed on the received data field synchronization code or a specified part thereof, and it is divided by the DFT of the standard data field synchronization code or a specified part thereof to determine the DTV DFT of the transmission channel. The DFT of the DTV transmission channel is normalized with respect to the maximum term to characterize the channel, and the Kernel weights of the digital equalization filter are selected to complement the normalized DFT characterizing the channel. For example, the No. 5331416 of C.B.Patel et al. published on July 19, 1994 entitled "METHODS FOR OPERATING GHOST-CANCELATION CIRCUITRY FOR TV RECEIVER OR VIDEORECORDER (the operation method of the ghost image elimination circuit of TV receiver or video recorder)" No. US Patent No. describes this adjustment method in more detail. This method is best used for the initial adjustment of the Kernel weight of the digital equalization filter, because this method can make the initial adjustment faster than the adaptive equalization. After the initial adjustment of the Kernel weight of the digital equalization filter, an adaptive equalization method is finally adopted. Awarded to J. Yang et al. on July 15, 1997 entitled "RAPID-UPDATE ADAPTIVE CHANEL-EQUALIZATION FILTERING FOR DIGITAL RADIO RECEIVERS, SUCH AS HDTV RECEIVERS (Rapid Update Adaptive Channel Equalization Filtering for Digital Radio Receivers and HDTV Receivers) "US Patent No. 5648987" describes a block LMS method for performing adaptive equalization. U.S. Patent Application No. 08/832,674, entitled "DYNAMICALLY ADAPTIVE EQUALIZER SYSTEM AND METHOD," filed April 1997 by A.L.R. Limberg, describes a continuous LMS method for performing adaptive equalization .

In step S9, especially when performing fast initial equalization before switching to adaptive equalization, DFT can be used to perform an auxiliary method to adjust the Kernel weight of the digital equalization filter, so that the cascaded digital equalization filter and comb filter The response of the filter is consistent with the standard response of this filter cascade. The adjustment is carried out as follows: Discrete Fourier Transform (DFT) is performed on the received data field synchronization code or a specified part thereof comb-filtered by the NTSC artifact suppression comb filter 20, using the standard data field synchronization code Or the DFT of the specified part which has been comb-filtered divides it, so as to determine the DFT of the DTV transmission channel. Then, the DFT of the DTV transmission channel is normalized with respect to the maximum term to characterize the channel, and the Kernel weights of the digital equalization filter are adjusted to complement the normalized DFT characterizing the channel. The adaptive equalization method is preferably used after the initial adjustment of the Kernel weights of the digital equalization filter. These adaptive equalization methods differ from those used when NTSC co-channel interference artifacts are small in that by using comb filter 20 to suppress NTSC artifacts, the possible effective signal states are doubled.

FIG. 4 is a schematic diagram showing partial details of the DTV signal receiver shown in FIG. 1, wherein the NTSC rejection comb filter adopts a structure of 120, and the post-coding comb filter adopts a structure of 126. Subtractor 1202 in NTSC rejection comb filter 120 acts as a first linear combiner and modulo-8 adder 1262 in post-coding comb filter 126 acts as a second linear combiner. The first delay device 1201 whose delay time is 12 symbol phase delays is used in the NTSC suppression comb filter 120, and the second delay device 1201 whose delay time is also 12 symbol phase delays is used in the post-coding comb filter 126 Timer 1263. The 12-symbol delay exhibited by each delay means 1201 and 1263 is approximately equal to the one-period delay of the analog TV video carrier artifact at a frequency 59.75 times the analog TV line scan frequency _fH . This 12 symbol delay is approximately equal to 5 periods of the analog TV chrominance subcarrier artifact at frequency _287.25fH . The 12 symbol delay is approximately equal to 6 periods of the analog TV sound carrier artifact at frequency _345.75fH . This is why the different combined responses of the subtractor 1202 to the audio carrier, to the video carrier and to those frequencies close to the differently delayed chrominance subcarriers of the first delay means 1201 will reduce co-channel interference. However, the amount of correlation in an analog TV video signal over such a horizontal spatial distance is very small in the portion of the video signal where the edges cross the horizontal scan lines.

When it is determined that there is not enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22, the multiplexer 1261 is controlled by the second state of the multiplexer control signal most of the time. When it is determined that there is enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22, the multiplexer 1261 is controlled by the first bit of the multiplexer control signal most of the time. Three-state control. Under the control of the third state of its control signal, the multiplexer 1261 feeds back the modulo-8 sum result delayed by the delay device 1263 for 12 symbol phase delays to the adder 1262 as the augend. This is a modular accumulation procedure in which a single error is propagated as a running error, cycled every 12 symbol phase delays. The multiplexer 1261 is in the first state during the 4-symbol phase delay at the beginning of each data segment and during the entirety of each data segment containing the field synchronization information, which reduces the post-coding output of the post-coding comb filter 126. The running error in the decoded result of an encoded symbol. When the control signal is in the first state, the output signal of the multiplexer 1261 reproduces the standard symbol decoding result of the output signal provided by the internal memory of the controller 28 . The output signal of multiplexer 1261 uses standard symbol decoding results to block running errors. Since there are 4+69(12) symbols per data segment, the standard decoding result is slipped back in phase by 4 symbol phase delays in each data segment, so it is unlikely that any running bit error will last longer than three data segments .

FIG. 5 is a detailed block diagram showing a version 144 of the NTSC co-channel interference detector 44 shown in FIG. Set 12 symbol delays with the Q channel signal of the equalization circuit 16 . The third linear combiner is a digital subtractor 1443, which generates a comb filter response that does not contain NTSC co-channel interference artifacts after differentially combining the symbol synchronization after differential delay and the Q channel signal output by the equalization circuit 16, The response signal is applied to amplitude detector 445 . The fourth linear combiner is a digital adder 1444 which superimposes and combines the differentially delayed Q-channel signals to produce a comb filter response containing selected NTSC co-channel interference artifacts which is added to the amplitude detection device 446. When the DTV signal receiver among Fig. 1 adopts the NTSC suppressing comb filter 20 of the kind 120 and the post-coding comb filter 26 of the kind of 126, this NTSC synchronous channel interference detector 144 is just particularly suitable for in Fig. 1 DTV signal receiver. Since the comb filtering using subtractor 1443 suppresses artifacts caused by the NTSC audio carrier, the NTSC video carrier and the NTSC color subcarrier, no bandwidth selection filter 441 is needed in front of node 440.

FIG. 6 is a schematic block diagram showing details of portions of the DVT signal receiver shown in FIG. 1 in the case of NTSC rejection comb filters 20 such as 220 and post-coding comb filters 26 such as 226 . The first delay device 2201 whose delay time is 6 symbol phase delays is used in the NTSC suppression comb filter 220, and the second delay device 2201 whose delay time is also 6 symbol phase delays is used in the post-coding comb filter 226. Delay device 2263. The 6-symbol delay presented by each delay means 2201 and 2263 is approximately equal to the 0.5 cycle delay of the analog TV video carrier artifact at a frequency 59.75 times the analog TV line scan frequency f _H , which is approximately equal to a frequency of 287.25 f _H 2.5 cycles of the analog TV chrominance subcarrier artifact, and approximately equal to 3 cycles of any artifact of the analog TV audio carrier at frequency 345.75f _H. Adder 2202 in NTSC rejection comb filter 220 acts as a first linear combiner and modulo-8 subtractor 2262 in post-coding comb filter 226 acts as a second linear combiner. Although the band of invalid frequencies (nulls) near the frequency converted from the analog TV carrier frequency is relatively narrow, the time delay presented by delay means 2201 and 2263 is smaller than the delay time presented by delay means 1201 and 1263, therefore, by The signals combined by the adder 2202 exhibit better anticorrelation than the signals differentially combined by the subtractor 1202 . The acoustic carrier suppression in the NTSC rejecting comb filter 220 response is poorer than the acoustic carrier rejection in the NTSC rejecting comb filter 120 response. However, the low sound rejection capability of the comb filter 220 is not a problem if the sound carrier in the co-channel interfering analog TV signal is suppressed by sound trap or SAW filtering in the IF amplifier circuit 12. In the case of using the NTSC rejection comb filter 220 of FIG. 6 instead of the NTSC comb filter 120 of FIG. The error correction function of structural decoding and Reed-Solomon decoding.

The multiplexer 261 of the type 2261 is controlled by the multiplexer control signal most of the time when it is determined that there is not enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22. The second state controls the multiplexer 261 of the type 2261 to be multiplexed most of the time when it is determined that there is enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22. The third state control of the user control signal. Under the control of the third state of its control signal, the multiplexer 2261 feeds back the modulo-8 sum result of the adder 2262 delayed by the delay device 2263 by 6 symbol phase delays to the adder 2262 as an augend. This is a modulo-accumulation process where single errors are propagated as running errors, cycled every 6 symbol phase delays. The multiplexer 2261 is in the first state during the 4-symbol phase delay at the beginning of each data segment and throughout the entire data segment of each data segment containing field sync information, which reduces post-coding comb filtering Running errors in the decoded result of the post-encoded symbols output by the unit 226. When this control signal is in the first state, the multiplexer 2261 reproduces as its output signal the standard symbol decoding result provided by the internal memory of the controller 28 . The output signal of the multiplexer 2261 uses the decoding result of the standard symbol to block the running error. Since each data segment has 4+138(6) symbols, in each data segment, the standard decoding result is the same as the 4-symbol phase delay, so the duration of any running bit error cannot exceed two data part. In the post-encoding comb filter 226, although running errors cycle faster and affect twice as many times as in the 12-interleaved trellis code, the possibility of long-term existence of running errors is lower than in the post-encoding comb filter 126 in the long-term existence of running error possibility.

7 is a detailed block diagram showing a version 244 of the NTSC co-channel interference detector 44 shown in FIG. 6 symbol delay. The third linear combiner is a digital adder 2443, which superimposes and combines the differentially delayed Q channel signals to produce a comb filter response without NTSC co-channel interference artifacts, which is added to the amplitude detection device 445. The fourth linear combiner is a digital subtractor 2444, which performs differential combination of the symbol synchronization after differential delay and the Q channel signal output by the equalization circuit to produce a comb filter response containing selected NTSC co-channel interference artifacts, which The response is applied to amplitude detector 446. When the DTV signal receiver among Fig. 1 adopts the kind of 226 post-coding comb filter 26 of NTSC rejection comb filter 20 of the kind 220, this NTSC co-channel interference detector 244 is just particularly suitable for in Fig. 1 DTV signal receiver.

FIG. 8 is a schematic diagram showing partial details of the DTV signal receiver shown in FIG. 1 in the case of using NTSC rejection comb filter 20 such as 320 and post-coding comb filter 26 such as 326 . The first delay device 3201 whose delay time is 1368 symbol phase delays has been used in the NTSC suppression comb filter 320, and the second delay device 3201 whose delay time is also 1368 symbol phase delays has been used in the post-coding comb filter 326. Delay device 3263, wherein the phase delay of 1368 symbols is approximately equal to the phase delay of two horizontal scanning lines of the analog TV signal. The first linear combiner in the NTSC rejection comb filter 320 is an adder 3202 and the second linear combiner in the post-coding comb filter 326 is a modulo-8 subtractor 3262 .

The multiplexer 261 of the type 3261 is most of the time controlled by the multiplexer control signal when it is determined that there is not enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22 The second state control, when it is determined that there is enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22, the multiplexer 261 of the type 3261 is subjected to the multiplexer most of the time. The third state control of the user control signal. The DTV signal receiver preferably includes detection circuitry for detecting the amount of change between interlaced lines of NTSC co-channel interference so that the controller 28 does not provide the third state of the multiplexer 32613261 control signal under these conditions.

Under the control of the third state of its control signal, the multiplexer 3261 feeds back the modulo-8 sum result of the adder 3262 delayed by the delay device 3263 for 1368 symbol phase delays to the adder 3262 as the augend. This is a modulo-accumulation process where single errors are propagated as running errors, cycled every 1368 symbol phase delays. The code length (span) of this symbol is larger than the length of a single Reed-Solomon code block, so a single running error is easily corrected in the Reed-Solomon decoding process. The multiplexer 3261 is in its first state during the 4-symbol phase delay at the beginning of each data segment and throughout the entire data segment of each data segment containing field sync information, which reduces post-coding comb Running errors in the decoded result of the post-encoded symbols output by filter 326 . When this control signal is in its first state, the multiplexer 3261 reproduces as its output signal the standard symbol decoding results provided by the internal memory of the controller 28. The output signal of the multiplexer 3261 uses the decoding result of the standard symbol to block the running error. An NTSC video field with a cycle time of 16.67 microseconds has a phase shift relative to a DTV data field with a cycle time of 24.19 microseconds, so a DTV data segment containing a field sync signal eventually scans the entire NTSC frame raster. The NTSC frame raster contains 525 lines, each line includes 684 symbol phase delays, for a total of 359100 symbol phase delays. Since this is slightly less than the 432 × 832 symbol phase delay contained in the DTV data segment containing the vertical sync signal, it is reasonable to speculate that running errors with a duration greater than 432 data fields will be reproduced during the DTV data segment containing the sync signal. The multiplexer 3261 of the decoding result of the standard symbol is now deleted. There is also a phase shift between the data segment where the standard symbol decoding result start block is available and the NTSC video scan line. It can be estimated that 359100 symbol phase delays can be scanned during 89775 consecutive data segments, which is 89775 times the 4 symbol phase delays contained in the initial block. Since there are 313 data segments in each DTV data field, it is reasonable to speculate that running errors with a duration greater than 287 data fields will be picked up by the multiplexer 3261 which reproduces the standard symbol decoding results during the initial block delete. These two sources of operational error suppression are independent of each other, so operational error fields with a duration greater than or equal to 200 data fields cannot exist. Also, if the NTSC co-channel interference is reduced when operating bit errors repeatedly occur so that the multiplexer 3261 reproduces the response of the data slicer 22 as its output signal, errors can be corrected earlier than would otherwise be the case. .

The NTSC rejection comb filter 320 shown in FIG. 8 does a good job of suppressing demodulation artifacts in response to the analog TV horizontal sync phase and a large portion of the demodulation artifacts in response to the analog TV vertical sync pulses and equalization pulses. Tune artifacts. These artifacts are co-channel interference with maximum energy. NTSC rejects the comb filter 320 regardless of the color of the video information, except in the case that the video information of the analog TV signal contains scam-line-to-scan-line variations between two line scans. The video information can be well suppressed. The FM audio carrier in the analog TV signal is well suppressed here if it is not suppressed by the tracking rejection filter in the symbol synchronization and equalization circuit 16 . In the response of the NTSC rejection comb filter 320, most of the artifacts of the analog TV color burst are also suppressed. Furthermore, the filtering by the NTSC rejection comb filter 320 is "orthogonal" to the NTSC interference rejection set into the trellis decoding process.

FIG. 9 is a detailed block diagram showing a version 344 of the NTSC co-channel interference detector 44 shown in FIG. 2 video line delays with 1368 symbol phase delays. The third linear combiner is a digital adder 3443, which superimposes and combines the differentially delayed Q channel signals to generate a comb filter response without NTSC co-channel interference artifacts, which is added to the amplitude detection device 445. The fourth linear combiner is a digital subtractor 3444, which performs differential combination of the symbol synchronization after differential delay and the Q channel signal output by the equalization circuit to generate a comb filter response containing selected NTSC co-channel interference artifacts, which The response is applied to amplitude detector 446. This NTSC co-channel interference detector 344 is particularly suitable for the DTV signal receiver of FIG. 1 when the DTV signal receiver of FIG. DTV signal receiver.

FIG. 10 is a schematic diagram showing some details of the DTV signal receiver shown in FIG. 1 in the case of using NTSC rejection comb filter 20 such as 420 and post-coding comb filter 26 such as 426 . The first delay device 4201 whose delay time is 179208 symbol phase delays has been used in the NTSC suppression comb filter 420, and the second delay device 4201 whose delay time is also 179208 symbol phase delays has been used in the post-coding comb filter 426. The delay device 4261, wherein, the phase delay of 179208 symbols is basically equal to 262 horizontal scanning line periods of the analog TV signal. The adder 4202 in the NTSC rejection comb filter 420 acts as a first linear combiner, and the modulo-8 subtractor 4262 in the post-coding comb filter 426 acts as a second linear combiner.

When it is determined that there is not enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22, the multiplexer 261 of the 4261 type is controlled by the multiplexer control signal most of the time. The second state control, when it is determined that there is enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22, the multiplexer 261 of the type 4261 is subjected to the multiplexer most of the time. The third state control of the user control signal. The DTV signal receiver preferably includes detection circuitry for detecting field-by-field changes in NTSC co-channel interference so that controller 28 can disable the third state of the control signal to multiplexer 4261 under these conditions.

Under the control of the third state of its control signal, the multiplexer 4261 feeds back the modulo-8 sum result of the adder 4262 delayed by the delay device 4263 by 179208 symbol phase delays to the adder 4262 as the augend. This is a modulo-accumulation process in which single errors are transmitted as running errors, cycled every 179208 symbol phase delays. This symbol code length is larger than the length of a single Reed-Solomon code block, so a single run error is easily corrected during Reed-Solomon decoding. The multiplexer 4261 is in the first state during the 4-symbol phase delay at the beginning of each data segment and throughout the entire data segment of each data segment containing the field sync signal, which reduces post-encoding comb filtering Running errors in the post-encoded symbol results output by the detector 426. When this control signal is in its first state, the multiplexer 4261 reproduces as an output signal the standard symbol decoding results provided by the internal memory of the controller 28 . The output signal of the multiplexer 4261 uses the decoding result of the standard symbol to block the running error. The maximum number of data fields required to eliminate running errors in the output signal of multiplexer 4261 is about the same as the maximum amount of data fields required to eliminate running errors in the output signal of multiplexer 3261. However, the number of bit errors during this period dropped to 1/131 of the original.

The NTSC rejection comb filter 420 shown in Figure 10 suppresses most of the demodulation artifacts in response to the analog TV vertical sync and equalization pulses and all demodulation in response to the analog TV horizontal sync pulses Artifacts. These artifacts are co-channel interference with maximum energy. The NTSC rejection comb filter 420 also suppresses artifacts caused by video information in the analog TV signal that does not vary from field to field or line to line, thereby eliminating static patterns that are not related to horizontal spatial frequency or color. In the NTSC rejection comb filter 420 response, most analog TV burst artifacts are also suppressed.

FIG. 11 is a detailed block diagram showing one form 444 of the NTSC co-channel interference detector 44 shown in FIG. Delay of 262 video lines (ie 179208 symbol phase delays). The third linear combiner is a digital adder 4443, which superimposes and combines the differentially delayed Q channel signals to generate a comb filter response without NTSC co-channel interference artifacts, which is added to the amplitude detection device 445. The fourth linear combiner is a digital subtractor 4444, which performs differential combination of the symbol synchronization after the differential delay and the Q channel signal output by the equalization circuit 16 to produce a comb filter response containing selected NTSC co-channel interference artifacts, The response is applied to magnitude detector 446 . This NTSC co-channel interference detector 444 is particularly suitable for the DTV signal receiver of FIG. 1 when the DTV signal receiver of FIG. DTV signal receiver.

FIG. 12 is a diagram showing partial details of the DTV signal receiver shown in FIG. 1 in the case of using the NTSC rejection comb filter 20 such as 520 and the post-coding comb filter 26 such as 526 . The first delay device 5201 whose delay time is 718200 symbol phase delays has been used in the NTSC suppression comb filter 520, and the second delay device 5201 whose delay time is also 718200 symbol phase delays has been used in the post-coding comb filter 526. The delay device 5261, wherein the delay of 718200 symbol phase delays is approximately equal to two frame periods of the analog TV signal. Subtractor 5202 in NTSC rejection comb filter 520 acts as a first linear combiner and modulo-8 adder 5262 in post-coding comb filter 526 acts as a second linear combiner.

When it is determined that there is not enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22, the multiplexer 261 of the 5261 type is controlled by the multiplexer control signal most of the time. The second state control, when it is determined that there is enough NTSC co-channel interference to cause uncorrectable bit errors in the output signal of the data slicer 22, the multiplexer 261 of the type 5261 is subjected to the multiplexer most of the time. The third state control of the user control signal. The DTV signal receiver preferably includes detection circuitry for detecting changes between alternate frames in NTSC co-channel interference so that the controller 28 can disable the third state of the control signal to the multiplexer 5261 under these conditions .

Under the control of the third state of its control signal, the multiplexer 5261 feeds back the modulo-8 sum result of the adder 5262 delayed by the delay device 5263 for 718,200 symbol phase delays to the adder 5262 as the augend. This is a modulo-accumulation process in which single errors are transmitted as running errors, cycled every 718200 symbol phase delays. This symbol code length is larger than the length of a single Reed-Solomon code block, so a single running code is easily corrected in the Reed-Solomon decoding process. The multiplexer 5261 is in the first state during the 4-symbol phase delay at the beginning of each data segment and throughout the entire data segment of each data segment containing the field sync signal, which reduces post-encoding comb filtering Running errors in the post-encoded symbol results output by the detector 526. When this control signal is in the first state, the multiplexer 5261 reproduces as its output signal the standard symbol decoding result provided by the internal memory of the controller 28 . The output signal of the multiplexer 5261 uses the decoding result of the standard symbol to block the running error. The maximum number of data fields required to eliminate running errors in the output signal of multiplexer 5261 is about the same as the maximum amount of data fields required to eliminate running errors in the output signal of multiplexer 3261. However, the number of bit errors during this period dropped to 1/525 of the original.

The NTSC rejection comb filter 520 shown in FIG. 12 suppresses all demodulation artifacts in response to the analog TV vertical sync pulses and equalization pulses, and suppresses all demodulation artifacts in response to the analog TV horizontal sync pulses. elephant. These artifacts are co-channel interference with maximum energy. The NTSC rejection comb filter 520 also suppresses artifacts caused by video information in the analog TV signal that does not change between two frames, thereby eliminating static patterns that are not related to their spatial frequency or color. All analog TV burst artifacts are also suppressed in the NTSC rejection comb filter 520 response.

FIG. 13 is a detailed block diagram showing a version 544 of the NTSC co-channel interference detector 44 shown in FIG. Set the delay of 2 video frames (ie 718200 symbol phase delay). The third linear combiner is a digital adder 5443, which superimposes and combines the differentially delayed Q channel signals to generate a comb filter response without NTSC co-channel interference artifacts, which is added to the amplitude detection device 445. The fourth linear combiner is a digital subtractor 5444, which performs differential combination of the symbol synchronization after the differential delay and the Q channel signal output by the equalization circuit 16 to produce a comb filter response containing selected NTSC co-channel interference artifacts, The response is applied to magnitude detector 446 . This NTSC co-channel interference detector 544 is particularly suitable for the DTV signal receiver of FIG. 1 when the DTV signal receiver of FIG. DTV signal receiver.

Those of ordinary skill in the art of television system design will recognize other properties of correlation and anticorrelation of analog TV signals that can be used in the design of NTSC rejection filters other than those shown in FIGS. Other types of NTSC rejection filters are utilized. The 2N levels of the baseband signal can be increased to (8N-1) data levels by using the NTSC suppression filter obtained by cascading two disclosed NTSC suppression filters, which increases the number of symbols in the decoding process. The difficulty of data clipping. Although these filters degrade the signal-to-noise ratio of random noise interference for symbol decoding, they can be used to specifically address severe co-channel interference problems. In this respect, cascaded filters for improving the NTSC rejection and selectivity of co-channel jammer detectors have few disadvantages.

FIG. 14 shows a cascade of filters for improving the NTSC rejection and selection capabilities of the co-channel jammer detector 44 as 644, which may be considered a modification of the co-channel jammer detector 544 shown in FIG. The response of FIR digital low pass filter 4410 to the equalized Q channel signal from symbol synchronization and equalization circuit 16 is applied at node 440. If the part of the comb filter that requires the maximum delay is placed at the front of the cascade, hardware can be saved because the same maximum delay element can be used by both the NTSC rejection comb filter and the NTSC selection comb filter. Same as the co-channel interference detector 544 in FIG. 13, in the co-channel interference detector 644 shown in FIG. 14, the signal on the node 440 is added to a delay element 5442 whose delay time is 2 video frames; The obtained differential delay signal is superposed and combined by the adder 5443 and differentially combined by the subtractor 5444 .

The sum response output by the adder 5443 is filtered by the next stage of NTSC suppression to generate the response of the NTSC suppression comb filter, which is added to the amplitude detector 445 . More specifically, the delay device 6441 sets a difference delay of 6 symbols for the sum response output by the adder 5443, and the sum response of the adder 5443 after the differential delay is superimposed and combined by the digital adder 6443 to generate The NTSC reject comb filter response is applied to the magnitude detector 445.

The difference response output by the subtractor 5444 is filtered by the next stage of NTSC suppression to generate the NTSC select comb filter response, which is applied to the amplitude detector 446 . More specifically, the delay device 6442 sets a difference delay of 6 symbols for the difference response output by the subtractor 5444, and the difference response of the subtractor 5443 after the difference delay is differentially combined by the digital subtractor 6444 to generate NTSC selects the comb filter response which is applied to magnitude detector 446. The amplitude detection results obtained by the amplitude detectors 445 and 446 are compared by the amplitude comparator 447 . Amplitude comparator 447 provides an output bit indicating whether the response of amplitude detector 446 substantially exceeds the response of amplitude detector 445 . This output bit is used to select the operating state (second state or third state) of the multiplexer 261 in the DTV signal receiver shown in Figure 1, here, the DTV signal receiver shown in Figure 1 adopts the same channel Interference detector 644 acts as its co-channel interference detector 44 . Cascaded filtering in co-channel interference detector 644 utilizes temporal comb filtering between alternating NTSC video frames to suppress NTSC artifacts caused by sync information and still video components. Cascaded filtering in co-channel interference detector 644 utilizes intra-frame spatial comb filtering to suppress NTSC artifacts caused by dynamic video components.

FIG. 15 further shows a cascade of filters for improving the NTSC rejection and selection capabilities of the co-channel jammer detector 44 as 744, which may be considered an improvement of the co-channel jammer detection 544 shown in FIG. The equalized Q-channel signal from the symbol synchronization and equalization circuit 16 is applied directly to node 440 and does not require FIR digital low pass filter. Same as the co-channel interference detector 544 in FIG. 13 , in the co-channel interference detector 744 shown in FIG. 15 , the signal on the node 440 is added to a delay element 5442 whose delay time is 2 video frames; The obtained differential delay signal is superposed and combined by the adder 5443 and differentially combined by the subtractor 5444 .

The sum response output by the adder 5443 should be filtered by the next stage of NTSC suppression to generate the response of the NTSC suppression comb filter, which is added to the amplitude detector 445 . More specifically, the delay device 7441 sets a difference delay of 12 symbols for the sum response output by the adder 5443, and the sum response of the adder 5443 after the differential delay is differentially combined by the digital subtractor 7443 to generate The NTSC reject comb filter response is applied to the magnitude detector 445.

The difference response output by the subtractor 5444 is filtered by the next stage of NTSC suppression to generate an NTSC suppression comb filter response, which is applied to the amplitude detector 446 . More specifically, the delay device 7442 sets a difference delay of 12 symbols for the difference response output by the subtracter 5444, and the difference response of the adder 5443 after the difference delay is superimposed and combined by the digital adder 7444 to generate NTSC selects the comb filter response which is applied to magnitude detector 446. The amplitude detection results obtained by the amplitude detectors 445 and 446 are compared by the amplitude comparator 447 . Amplitude comparator 447 provides an output bit indicating whether the response of amplitude detector 446 substantially exceeds the response of amplitude detector 445 . This output bit is used to select the operating state (second state or third state) of the multiplexer 261 in the DTV signal receiver shown in Figure 1, here, the DTV signal receiver shown in Figure 1 adopts the same channel The interference detector 744 acts as its co-channel interference detector 44 . Cascaded filtering in co-channel interference detector 744 utilizes temporal comb filtering between alternate NTSC video frames to suppress NTSC artifacts caused by sync information and still video components. Cascaded filtering in co-channel interference detector 744 utilizes intra-frame spatial comb filtering to suppress NTSC artifacts caused by audio components and dynamic video components.

Figure 16 shows a modification of the DTV signal receiver shown in Figure 1 described thus far, constructed in accordance with another aspect of the present invention, which employs a plurality of even-level data slicers A24, A24, B24 and C24, each even-level data limiter is preceded by a corresponding NTSC suppression comb filter, followed by a corresponding post-coding comb filter. The even-level data slicer A24 converts the response of the NTSC reject filter A20 of the first type into first precoded symbol decoding results for use by the post-code comb filter A26 of the first type. The even-level data slicer B24 converts the response of the second type of NTSC rejection filter B20 into second pre-coded symbol decoding results for use by the second type of post-coding comb filter B26. The even-level data slicer C24 converts the response of the second-type NTSC rejection filter C20 into third-precoded symbol decoding results for use by a third-type post-coding comb filter C26. Odd level data slicer 22 provides intermediate symbol decoding results to post-coding comb filters A26, B26 and C26. When the receiver part adopts some of the parts shown in Figures 4, 6, 8, 10 and 12, the prefixes A, B and C of the part numbers in Figure 15 correspond to the integers 1, 2, 3, 4 and 5, respectively different integers.

The first-type co-channel interference detector A44 determines from the Q-channel signal the effectiveness of the first-type NTSC rejection filter A20 to suppress co-channel interference from the analog TV signal in the currently equalized I-channel signal. The second-type co-channel interference detector B44 determines from the Q-channel signal the effect of the second-type NTSC rejection filter B20 on co-channel interference suppression from the analog TV signal in the currently equalized I-channel signal. A third type of co-channel interference detector C44 determines from the Q channel signal the effect of the third type of NTSC rejection filter C20 on co-channel interference suppression from the analog TV signal in the currently equalized I channel signal. The suppression of the pilot carrier in the Q channel signal helps the co-channel interference detectors A44, B44 and C44 provide an indication of the relative suppression effectiveness of the NTSC rejection comb filters A20, B20 and C20.

The symbol decode selection circuit 90 generates corrected symbol decode best estimates for use by the data assembler 30 . This best estimate is obtained by post-coding the standard symbol decoding results from the controller 28, the intermediate symbol decoding results from the odd-level data slicer 22, and the post-coding comb filters A26, B26, and C26. Symbol decoding results are selected and generated. Unless the controller 28 provides other symbol selection information to the symbol decoding selection circuit 90, the symbol decoding selection circuit 90 will determine this best estimate based on the suppression effectiveness indications provided by the co-channel interference detectors A44, B44 and C44. Other symbol selection information provided by the controller 28 includes indication information of the occurrence time of the synchronization code, which is used to ensure that the best estimated value is obtained on the basis of the standard symbol decoding result provided by the controller 28 . The best estimate of the symbol decoding result is used to correct the summing operation of the matched comb filters A26, B26 and C26 in the preferred embodiment of the DTV signal receiver of FIG. 16.

Symbol decode selection circuit 90 selects the middle symbol from odd-level data slicer 22 if co-channel jammer detectors A44, B44, and C44 all indicate the absence of major NTSC sync jamming artifacts outside of sync field occurrences The decoding result is used as the best estimate of the decoding result of the correction symbol. This reduces the impact of Johnson noise on symbol decoding.

If at least one of the co-channel interference detectors A44, B44 and C44 indicates a large NTSC sync interference artifact outside of the sync code occurrence, the symbol decode selection circuit 90 selects one of the NTSC rejection comb filters A20, Decoded results of post-coded symbols output by post-coded comb filters A26, B26, or C26 behind B20 and C20, said certain NTSC-suppressing comb-filter being determined to be NTSC by co-channel interference detectors A44, B44, and C44 The NTSC rejection comb filter that best suppresses co-channel interference artifacts.

All high-energy demodulation artifacts in response to analog TV sync pulses, equalization pulses, and color bursts are suppressed when the NTSC rejection comb filter A20 overlays and combines alternate video frames. Artifacts caused by video information in an analog TV signal that does not change between frames are also suppressed, thereby eliminating static patterns that are not related to their spatial frequency or color. A co-channel interference detector A44 of the type shown in FIG. 13 is used with a symbol decoding circuit of the type shown in FIG.

The remaining problem of suppressing demodulation artifacts concerns suppressing those demodulation artifacts caused by frame-to-frame differences at certain pixel positions within the analog TV signal grid. These demodulation artifacts can be suppressed using intra-frame filtering techniques. Optional NTSC rejection comb filter B20 and post-encoding comb filter B26 circuit utilizes correlation in the horizontal direction to suppress remaining demodulation artifacts, optional NTSC rejection comb filter C20 and post-encoding comb filter C26 The circuit uses the correlation in the vertical direction to suppress the remaining demodulation artifacts. Other possible implementation methods of this design idea are considered below.

If the voice carrier in the co-channel interfering analog TV signal is not suppressed by SAW filtering or by the voice trap circuit in the IF amplifier circuit 12, it is convenient to combine the NTSC rejection comb filter B20 with the post-coding comb filter B26 The circuit is selected as the NTSC rejection comb filter 120 and post-coding comb filter 126 circuits shown in FIG. 4 . A co-channel interference detector B44 of the type shown in FIG. 5 is used with a symbol decoding circuit of the type shown in FIG.

If the sound carrier in the synchronous interference analog TV signal is suppressed with a SAW filter circuit or with a sound trap circuit in the IF amplifier circuit 12, the NTSC suppression comb filter B20 and the post-coding comb filter B26 circuit can be easily combined The NTSC rejection comb filter 220 and post-coding comb filter 226 circuits of the kind shown in FIG. 6 are chosen. This is because the anticorrelation between video components that are only 6 symbol phase delayed from each other is generally better than the correlation between video components that are several symbol phase delayed from each other. A co-channel interference detector B44 of the type shown in FIG. 7 is used with a symbol decoding circuit of the type shown in FIG.

Optimal selection of the NTSC rejection comb filter C20 and postcoding comb filter C26 circuits is not an easy task, since the selection (taking into account the field spacing in the interfering analog TV signal) of the NTSC rejection comb In the shape filter C20, it is selected whether to combine the temporally closer scan line with the current scan line in the same field or to use the spatially closer scan line in the previous field to combine with the current scan line. It is usually a better choice to pick temporally closer scan lines in the same field, since jump cuts between fields are less likely to impair the NTSC suppression effect of the comb filter C20. For this option, the NTSC rejection comb filter C20 and post-encoding comb filter C26 circuits would use the NTSC rejection comb filter 320 and post-encoding comb filter 326 circuits shown in FIG. A co-channel interference detector C44 of the type shown in FIG. 9 is used with a symbol decoding circuit of the type shown in FIG.

Alternatively, the NTSC rejection comb filter C20 and post-encoding comb filter C26 circuits would use the NTSC rejection comb filter 420 and post-encoding comb filter 426 circuits shown in FIG. A co-channel interference detector C44 of the type shown in FIG. 11 is used with a symbol decoding circuit of the type shown in FIG.

Claims (19)

1. method that is used to handle the vestigial sideband amplitudemodulation digital television signal of digital television signal receiver said method comprising the steps of:

The vestigial sideband amplitudemodulation digital television signal that is subject to co-channel NTSC interference effect is carried out compound demodulation, so as to isolate the I channel baseband signal that received and with the Q channel baseband signal that is received of the described orthogonal thereto relation of I channel baseband signal that receives; With

Whether other artefact that disturbs by the co-channel NTSC that determines to follow in the described Q channel baseband signal that receives surpasses specified level, and whether the artefact that the co-channel NTSC that estimates to follow in the described I channel baseband signal that receives disturbs reaches significant level.

2. whether one kind be used to determine to adopt before the lattice decoding of digital television signal receiver co-channel NTSC to disturb the method that suppresses comb filtering, said method comprising the steps of:

The vestigial sideband amplitudemodulation digital television signal is carried out compound demodulation, so as to isolate the I channel baseband signal that received and with the Q channel baseband signal that is received of the described orthogonal thereto relation of I channel baseband signal that receives;

Determine whether to follow in the described Q channel baseband signal that receives co-channel NTSC to disturb artefact with significant level;

If determine not exist in the described Q channel baseband signal that receives described co-channel NTSC to disturb artefact with significant level, then the described I channel baseband signal that receives of not passing through comb filtering is carried out symbol decoding, thereby produce the decoded symbol of using for described lattice decoding;

If determine to exist in the described Q channel baseband signal that receives described co-channel NTSC to disturb artefact with significant level, just the described I channel baseband signal that receives is carried out comb filtering, thus the I channel baseband signal that comb filtering is crossed produced;

If determine to exist in the described Q channel baseband signal that receives described co-channel NTSC to disturb artefact, just the I channel baseband signal that described comb filtering is crossed is carried out symbol decoding with significant level; With

If determine to exist in the described Q channel baseband signal that receives described co-channel NTSC to disturb artefact with significant level, just the symbol decoding result of the I channel baseband signal that described comb filtering is crossed carries out the back coding, thereby produces the decoded symbol of using for described lattice decoding.

3. method according to claim 2, further comprising the steps of:

If determine to exist in the described Q channel baseband signal that receives described co-channel NTSC to disturb artefact with significant level, just before the I channel baseband signal that comb filtering is crossed was advanced described symbol decoding, it is consistent with the response of standard comb filter that described comb filtering is crossed the response of described I channel baseband signal; With

If determine to exist in the described Q channel baseband signal that receives described co-channel interference artefact with significant level, just before the described I channel baseband signal that receives of not passing through comb filtering is carried out described symbol decoding, the described I channel signals that receives that does not pass through comb filter is carried out equilibrium.

4. digital television signal receiver comprises:

Amplifier circuit is used to provide the vestigial sideband amplitudemodulation digital television signal of amplification of co-channel interference simulation TV signal of being easy to carry about with one;

Compound demodulator, vestigial sideband amplitudemodulation digital television signal after the described amplification is responded, so that I channel baseband signal that contains any co-channel interference simulation TV signal artefact and the Q channel baseband signal that contains other artefact of any co-channel interference simulation TV signal to be provided;

The symbol decoding device of described I channel baseband signal, comprise first data amplitude limiter that is used in the very first time, described I channel baseband signal being carried out symbol decoding, usually as long as the described artefact of any co-channel interference simulation TV signal is lower than the significant level of described I channel baseband signal, the error code among the first symbol decoding result of then described first data amplitude limiter output can be proofreaied and correct; With

Described Q channel baseband signal is played the co-channel interference detector of NTSC of response, those of described other artefact that are used for detecting any co-channel interference simulation TV signal are higher than the appearance of the artefact of described Q channel baseband signal significant level, and the described significant level of described Q channel baseband signal is consistent with the described significant level of described I channel baseband signal.

5. digital television signal receiver according to claim 4, wherein, the co-channel interference detector of described NTSC that described Q channel baseband signal is played response comprises:

Time-delay mechanism, the described Q signal baseband signal that is used to delay time is to produce the Q channel baseband signal of difference time-delay;

Adder is used for the Q channel baseband signal after the described difference time-delay is carried out stack combinations, to produce and value signal;

Subtracter is used for the Q channel baseband signal after the described difference time-delay is carried out the difference combination, to produce difference signal;

First amplitude detector is used to detect described and amplitude value signal, detects response to produce first amplitude;

Second amplitude detector is used to detect the amplitude of described difference signal, detects response to produce second amplitude; With

Magnitude comparator, be used for more described first and second amplitude and detect response, and show that described other artefact of any co-channel interference simulation TV signal is higher than the described significant level of described Q channel baseband signal during greater than setting when the difference of described first and second amplitude response.

6. digital television signal receiver according to claim 5, wherein, the described time-delay mechanism in the co-channel interference detector of described NTSC produces the Q channel baseband signal after the described difference time-delay with the time-delay of 12 symbol phase delay differences.

7. digital television signal receiver according to claim 5, wherein, the described time-delay mechanism in the co-channel interference detector of described NTSC produces the Q channel baseband signal after the described difference time-delay with the time-delay of 6 symbol phase delay differences.

8. digital television signal receiver according to claim 5, wherein, the described time-delay mechanism in the co-channel interference detector of described NTSC produces the Q channel baseband signal after the described difference with 1368 symbol phase delays or 2 ntsc video scan line difference time-delays is delayed time.

9. digital television signal receiver according to claim 5, wherein, the described time-delay mechanism in the co-channel interference detector of described NTSC produces the Q channel baseband signal after the described difference with 179208 symbol phase delays or 262 ntsc video scan line difference time-delays is delayed time.

10. digital television signal receiver according to claim 5, wherein, the described time-delay mechanism in the co-channel interference detector of described NTSC produces the Q channel baseband signal after the described difference with 718200 symbol phase delays or 2 ntsc video frame difference time-delays is delayed time.

11. digital television signal receiver according to claim 4, wherein, the co-channel interference detector of described NTSC that described Q channel baseband signal is played response comprises:

First time-delay mechanism, the described Q channel baseband signal that is used to delay time is to produce the Q channel baseband signal after difference is delayed time;

First adder is used for the Q channel baseband signal after the time-delay of described difference is carried out stack combinations, to produce first and value signal;

First subtracter is used for the Q channel baseband signal after the described difference time-delay is carried out the difference combination, to produce first difference signal;

Second time-delay mechanism is used to delay time described first and value signal, to produce first and value signal after the difference time-delay;

The 3rd time-delay mechanism, described first difference signal is used to delay time, producing first difference signal after the difference time-delay, the method that described the 3rd time-delay mechanism is delayed time to described first difference signal is similar to described second time-delay mechanism to described first and the value signal method of delaying time;

Second adder is used for first and value signal after the time-delay of described difference are carried out stack combinations, to produce second and value signal;

Second subtracter is used for first difference signal after the described difference time-delay is carried out the difference combination, to produce second difference signal;

First amplitude detector is used to detect described second and the amplitude of value signal, detects response to produce first amplitude;

Second amplitude detector is used to detect the amplitude of described second difference signal and produces second amplitude and detects response; With

Magnitude comparator, be used for that described first amplitude is detected response and detect response with second amplitude and compare, and when described first amplitude detects the difference of response between responding with the detection of second amplitude greater than setting, show that described other artefact of any co-channel interference simulation TV signal is higher than the described significant level of described Q channel baseband signal.

12. digital television signal receiver according to claim 4, wherein, the co-channel interference detector of described NTSC that described Q channel baseband signal is played response comprises:

First time-delay mechanism, the described Q channel baseband signal that is used to delay time is to produce the Q channel baseband signal after difference is delayed time;

First adder is used for the Q channel baseband signal after the time-delay of described difference is carried out stack combinations, to produce first and value signal;

First subtracter is used for the Q channel baseband signal after the described difference time-delay is carried out the difference combination, to produce first difference signal;

Second time-delay mechanism is used to delay time described first and value signal, to produce first and value signal after the difference time-delay;

The 3rd time-delay mechanism, described first difference signal is used to delay time, producing first difference signal after the difference time-delay, the method that described the 3rd time-delay mechanism is delayed time to described first difference signal and described second time-delay mechanism are identical with the value signal method of delaying time to described first;

Second adder is used for first difference signal after the described difference time-delay is carried out stack combinations, to produce second difference signal;

Second subtracter is used for first and value signal after the time-delay of described difference are carried out the difference combination, to produce second and value signal;

First amplitude detector is used to detect described second and the amplitude of value signal, detects response to produce first amplitude;

Second amplitude detector is used to detect the amplitude of described second difference signal, detects response to produce second amplitude; With

Magnitude comparator, be used for that described first amplitude is detected response and detect response with second amplitude and compare, and when described first amplitude detects the difference of response between responding with the detection of second amplitude greater than setting, show that described other artefact of any co-channel interference simulation TV signal is higher than the described significant level of described Q channel baseband signal.

13. digital television signal receiver according to claim 4 also comprises:

Data synchronization circuit is used for the specified data synchronizing symbol in moment that described I channel baseband signal occurs; With

Be used for occurring the circuit of the moment generation standard symbol decoded result of data sync symbol in described I channel baseband signal; Wherein said sign coding equipment also comprises:

First time-delay mechanism produces the time-delay that is used for described I channel baseband signal is risen first defined amount symbol phase delay of response, thereby produces the I channel baseband signal after the difference time-delay;

First linear combiner carries out linear combination to the I channel baseband signal after the described difference time-delay, and to produce the response of first comb filter, in this response, described any co-channel interference anolog TV signals are inhibited;

Second data amplitude limiter is used for symbol decoding is carried out in described first comb filter response, so that produce the symbol decoding result after first precoding;

Second linear combiner, corresponding first and second input signal that receives is carried out linear combination, thereby provide corresponding output signal to respond as second comb filter, described second linear combiner is used to receive symbol decoding result after described first precoding as corresponding first input signal, have one to be adder in described first linear combiner and described second linear combiner, another then is a subtracter;

Second time-delay mechanism is used for its respective input signals described first defined amount symbol phase of delaying time is postponed, so that produce described second input signal of described second linear combiner;

Input multiplexer more than first, be used for its corresponding output signal is offered described second input signal of described second time-delay mechanism as it, be used to receive described standard symbol decoded result as its first input signal, be used to receive described intermediate symbols decoded result as its second input signal, and be used to receive the described output signal of described second linear combiner as its 3rd input signal, when the data sync symbol occurring in definite described I channel baseband signal that and if only if, described first multiplexer reappears its first input signal as its output signal, when other artefact that detects described any co-channel interference simulation TV signal when the co-channel interference detector of described NTSC is higher than the described significant level of described Q channel baseband signal, described first multiplexer reappears the output signal of described second linear combiner as its output signal, when other artefact that does not detect described any co-channel interference simulation TV signal when the co-channel interference detector of described NTSC was higher than the described significant level of described Q channel baseband signal, described first multiplexer reappeared the output signal of described first data amplitude limiter as its output signal.

14. digital television signal receiver according to claim 13, wherein, the co-channel interference detector of described NTSC that described Q channel baseband signal is played response comprises:

The 3rd time-delay mechanism is used for described Q channel baseband signal first defined amount the symbol phase of delaying time is postponed, so that produce the Q channel baseband signal after the difference time-delay;

The trilinear combiner, with the Q channel baseband signal linear combination after the described difference time-delay, to produce the response of the 3rd comb filter, in this response, the artefact of any co-channel interference simulation TV signal all is tending towards being inhibited;

The 4th linear combiner, with the Q channel baseband signal linear combination after the described difference time-delay, to produce the response of the 4th comb filter, in this response, the artefact of any co-channel interference simulation TV signal all is tending towards being enhanced, have one to be adder in described trilinear combiner and described the 4th linear combiner, another then is a subtracter;

First amplitude detector is used to detect the amplitude that described the 3rd comb filter responds, and detects response to produce first amplitude;

Second amplitude detector is used to detect the amplitude that described the 4th comb filter responds, and detects response to produce second amplitude; With

Magnitude comparator, be used for that described first amplitude is detected response and detect response with second amplitude and compare, when the difference between described first amplitude response and second amplitude response shows that described any co-channel other artefact of interference simulation TV signal is higher than the described significant level of described Q channel baseband signal during greater than setting.

15. digital television signal receiver according to claim 14, wherein, described first, second delayed time with the difference that the 3rd time-delay mechanism all provides 12 symbol phase to postpone; Described first and the trilinear combiner be subtracter; The described second and the 4th linear combiner is an adder.

16. digital television signal receiver according to claim 14, wherein, described first, second delayed time with the difference that the 3rd time-delay mechanism all provides 6 symbol phase to postpone; Described first and the trilinear combiner be adder; The described second and the 4th linear combiner is a subtracter.

17. digital television signal receiver according to claim 14, wherein, described first, second and the 3rd time-delay mechanism all provide 1368 symbol phase to postpone or 2 ntsc video scan lines the difference time-delay; Described first and the trilinear combiner be adder; The described second and the 4th linear combiner is a subtracter.

18. digital television signal receiver according to claim 14, wherein, described first, second all provides 179208 symbol phase to postpone with the 3rd time-delay mechanism or the difference of 262 ntsc video scan lines is delayed time; Described first and the trilinear combiner be adder; The described second and the 4th linear combiner is a subtracter.

19. digital television signal receiver according to claim 14, wherein, described first, second all provides 718200 symbol phase to postpone with the 3rd time-delay mechanism or the difference of 2 ntsc video frames is delayed time; Described first and the trilinear combiner be adder; The described second and the 4th linear combiner is a subtracter.

Images