MXPA98005803A - T.v. digital receiver circuit to detect and delete co interchange interference - Google Patents

Abstract

The present invention relates to a digital television signal receiver, characterized in that it comprises: an apparatus for the detection of digital television signals for supplying a stream of symbols of level 2N each having a time of symbols of a specific length in the time, where N is a positive integer, said symbol current of level 2N is capable of being accompanied by analog signal artifacts of co-channel interference television, said symbols being grouped in the successive data segments with respective headers of the code of synchronization of data segments, the data segments are grouped into successive data fields with the initial data segment of each data field, which contains a data field synchronization code which changes from data field to data field The set of circuits for the provision of an M number of unique filter responses in comb ot When combing said 2N level symbol current, each comb filter-type response is less likely to be accompanied by artifacts of the co-channel interference television analog signal than the 2N level symbol current; symbols for generating decoding results of respective estimated symbols, a first of the plurality of symbol decoders generates first estimated symbol decoding results, which respond to said 2N level symbol stream, each other of the plurality of symbol decoders generates decoding results of respective, estimated symbols for a respective one of the comb-like filter M responses, whose respective estimated symbol decoding results are post-encoded to complete the respective balanced filtering for a respective one of the M unique comb filter responses. a pa Each of the plurality of symbol decoders includes a second symbol decoder for generating second estimated decoding results of symbols, the set of circuits for detecting whether the resulting symbol decoding data is estimated, respectively. presently there is an exit or departure between the first and second decoding results of symbols, estimated, and the set of circuits for selecting the best estimate, to select the best estimates from the symbol decoding results, estimated, respectively, to generate final decoding results of symbols at times between those times when the synchronization codes occur, the selection of the best estimates depends on the outputs from the first symbol decoding results, estimates of other symbol decoding results, estimate

Description

- SET OF DIGITAL TV RECEIVER CIRCUITS TO DETECT AND DELETE NTSC CO-CHANNEL INTERFERENCE FIELD OF THE INVENTION The present invention relates to digital television systems, such as the high definition digital television (HDTV) system used for terrestrial broadcasting in the United States.

United States of America in accordance with the standard of the Advanced Television Subcommittee (Advanced Television Sub-Co mi tt ee (ATSC)), and more particularly, to digital television receivers with circuit assemblies to detect co-channel interference to from analogue television signals, such as those that conform to the standard of the National Television Systems Committee (NTSC).

BACKGROUND OF THE INVENTION A Digital Television standard published on September 16, 1995 by the Subcommittee of Advanced Television (ATSC) specifies the signals of residual sideband (VSB) for the transmission of REF: 27968 Digital television (DTV) signals on 6 MHz bandwidth television channels, such as those currently used in the air broadcast of analog television signals of the National Television Subcommittee (NTSC) within the United States . The VSB DTV signal is designed so that its spectrum is likely to be interspersed with the spectrum of an NTSC analog television signal of co-channel interference. This is done by placing the pilot carrier and the sideband frequencies of the main amplitude modulation of the DTV signal at odd multiples of a quarter the horizontal scanning line ratio of the NTSC analog TV signal that falls within the multiple pairs of a quarter the horizontal scan line ratio of the NTSC analog TV signal, at whose even multiples the most part of the energy of the luminance and chrominance components of an analog NTSC TV signal will fall co-channel interference. The video carrier of the NTSC analog TV signal is shifted 1.25 MHz from the lower limit frequency of the television channel. The carrier of the DTV signal is displaced from such a video carrier by 59.75 times * # the horizontal scan line ratio of the NTSC analog TV signal, to place the DTV signal carrier at approximately 309.877.6 kHz from the lower limit frequency of the television channel. Accordingly, the carrier of the DTV signal is approximately 2 '690, 122.4 Hz of the intermediate frequency of the television channel. The exact symbol ratio in the Digital Television Standard is (684/285) times the displacement of the 4.5 MHz sound bearer of the video carrier on an analog TV signal of NTSC. The number of symbols per horizontal scanning line in an NTSC analog TV signal is 684, and 286 is the factor by which the ratio of horizontal scan line in an NTSC analog TV signal, to obtain the displacement of the 4.5 MHz sound bearer from the video carrier in an NTSC analog TV signal. The proportion of symbols is 10.762238 egasymbols per second, which may be contained in a VSB signal extending 5.381119 MHz from the DTV signal carrier. That is, the VSB signal may be limited to a band that extends 5.690997 MHz from the frequency lower limit of the television channel.

^ The ATSC standard for terrestrial broadcasting of HDTV digital signals in the United States of America is capable of transmitting any of two formats of high definition television (HDTV) with a 16: 9 aspect ratio. An HDTV screen format uses 1920 samples per scan line and 1080 active horizontal scan lines per 30 Hz frame with a # Field interlacing of 2: 1. The other format of HDTV screen uses 1280 luminance samples per scan line and 720 scan lines progressively scanned from the television picture per frame or 60 Hz frame. The ATSC standard also accommodates the transmission of the formats of DTV screen different from the HDTV screen formats, such as the parallel transmission of four television signals that have normal definition compared to an analog TV signal NTSC. 20 The DTV transmitted by residual sideband amplitude modulation (VSB) during terrestrial broadcasting in the United States of America, comprises a succession of consecutive data fields in time, each containing 313 consecutive data segments in time. The * t data fields can be considered as consecutively numbered in 2 modules, with each data field numbered with odd numbers and the subsequent data field numbered with numbers 5 pairs, forming a data frame. The proportion or speed of the frame is 20.66 frames per second. Each data segment is 77.3 microseconds of y. || duration. Thus, with the speed or proportion of symbols that is 10.76 MHz there are 832 symbols per data segment. Each data segment begins with a group of line synchronization code of four symbols, which have successive values of + S, -S, -S and + S. The value + S is one level below the excursion of the maximum positive data, and the value -S is a level above the excursion of the maximum negative data. The initial line of each data field includes a group of field synchronization code that encodes a training signal or instruction for matching procedures or channel compensation and multiple trajectory suppression. The instruction signal is a pseudo-noise sequence of 511 samples (or "PN sequence") followed by three PN sequences of 63 samples. Intermediates of PN sequences of 63 samples in field synchronization codes are * transmitted according to a first logical convention on the first line of each data field numbered with odd numbers and according to a second logical convention on the first line of each data field numbered with even numbers, the first and second logical conventions which are respective and complementary to each other. The data within the data lines are # encoded by trellis codes using twelve interleaved trellis codes, each of the trellis codes of 2/3 ratio with an uncoded bit. Interleaved trellis codes are subject to Reed-Solomon's forward error correction coding, which provides correction for chromatic timing errors arising from noise sources such as a nearby, nearby automobile ignition system. The coding results by Reed-Solomon are transmitted as coding of constellation symbols one-dimensional of 8 levels (3 bits / symbol) for transmission over air, whose transmissions are made without precoding the separate symbol of the trellis coding procedure. The results of the Reed-Solomon coding are transmitted as the coding of symbols of One-dimensional constellation 16 levels (4 bits / symbol) for cable broadcast, whose transmissions are made without precoding. The VSB signals have their natural carrier wave, which can vary in amplitude depending on the modulation percentage, suppressed. The natural carrier wave is replaced by a fixed amplitude pilot carrier wave, whose ^ amplitude corresponds to a prescribed percentage of modulation. This fixed amplitude pilot carrier wave is generated by the introduction of a direct component displacement in the modulation voltage applied to the balanced modulator, which generates the amplitude modulation sidebands that are supplied to the filter that provides the VSB signal as its response. If the eight levels of 4-bit symbol coding have normalized values of -7, -5, -3, -1, + 1, +3, +5 and +7 in the carrier modulation signal, the pilot carrier has a normalized value of 1.25. The normalized value of + S is +5, and the normalized value of -S is -5. In the first developments of the DVT technique it was contemplated that the DTV broadcaster can be called to decide whether or not to use a * symbol precoder in the transmitter, whose symbol precoder would follow the set of symbol generation circuits and provide precoded symbol filtering. This decision by the broadcaster would depend on whether the interference from an NTSC co-channel broadcast station was expected or not. The symbol precoder would complement the post¬ # symbol coding introduced incidentally in each DTV receiver by a comb characteristic filter used before the data switch in the symbol decoder circuitry to reject the artifacts of the co-channel interference signal NTSC. The precoding of symbols would not have to i ?? be used for the data line synchronization code groups or during the data lines in which the synchronization data in the data field was transmitted. 20 Co-channel interference is reduced to greater distances from the NTSC broadcast station (s) and is more likely to occur when certain ionospheric conditions are obtained, the summer months during the high activity years solar that are notorious for the probability of * Co-channel interference. Such interference will not occur if there are no co-channel NTSC broadcast stations, of course. If there were a probability of NTSC interference within its broadcast coverage area, it would be presumed that the HDTV broadcaster would use the symbol precoder to facilitate the HDTV signal being more easily separated from NTSC interference; and, consequently, a comb characteristic filter would be employed as a post-encoder of symbols in the DTV receiver to complete the adapted or balanced filtering. If there is no possibility of NTSC interference or there is only an insubstantial probability of it, in order to Since this noise has a flat frequency spectrum less likely to cause erroneous decisions for the symbol values in the trellis decoder, it was presumed that the DTV diffuser could discontinue the use of the symbol precoder; and in Consequently, the post-coding of symbols would then be disabled in each DTV receiver. Without the broadcaster or broadcaster who is aware of the condition, the NTSC or channel interference may be substantial for portions of the reception area for a broadcast, due to the jump conditions * erroneous, due to the leakage of the cable broadcast, due to the suppression of inadequate intermediate frequency image in NTSC receivers, due to the magnetic tape used for the digital recording of television that has analogical recording of previous television, remnant, or due to some other unusual condition. U.S. Patent No. 5,594,496 issued January 14, 1997 to Nielsen and collaborators entitled "DETECTION OF CO-CHANNEL INTERFERENCE IN DIGITAL TELEVISION SIGNALS" describes an NTSC co-channel interference detector in which data segments containing field synchronization codes are subject to filtering in comb of the data field to obtain a data field comb filter response with periodic free symbol coding intervals, in which co-channel interference and received noise can be evaluated. The response of the comb filter of the data field during these periodic intervals, is compared to that response as subject to the additional comb filtering that suppresses the artifacts of the NTSC co-channel interference. If the additional comb filtration gives as result the appreciable reduction in the level of signal, it is presumed that the signal comprises substantial artifacts of the co-channel interference of NTSC, and the comb filtering to suppress the artifacts of the NTSC co-channel interference is employed before the decoding of symbols. If the additional comb filtering does not result in appreciable reduction in detected noise, it is presumed that the noise is primarily Johnson-type noise, and the • Decoding of symbols is used without using previously comb filtering to suppress artifacts from NTSC co-channel interference. This is because comb filtering in which differentially delayed symbol coding is linearly combined is associated with a 3 dB or similar increase in Johnson noise. The current ATSC DTV standard does not authorize the transmitter to use symbol precoding. The suppression of the analog TV signal of co-channel interference is presumed to be carried out in the trellis type decoding process, after the data sectioning procedures, associated with the decoding of symbols. This procedure avoids the problem of determining if precoding is or is not done in the transmitter. However, the Analog TV signal of co-channel interference, undesirably introduces errors in the data sectioning processes, which places more burden on the procedures for decoding error correction, trellis-type decoding, and Reed-type decoding -Solomon. These errors will reduce the broadcast coverage area, which may lose revenue for the commercial DTV broadcaster. Thus, the provision for the suppression of the co-channel analogue TV signal before the data sectioning is still desirable, in spite of the precoding of symbols in the DTV transmitter that is not authorized by the DTV standard. of current ATSC. The term "linear combination" refers generically to addition and subtraction, performed either in accordance with conventional arithmetic or modular arithmetic. The term "modular combination" refers to the linear combination carried out according to a modular arithmetic. That type of coding that recodes a stream of digital symbols through differential delay and the linear combination of differentially delayed terms, ^ exemplified by the post t-coding of symbols used in the HDTV receivers of the prior art, is defined as the "recoding of symbols of the first type" in this specification. That type of coding that recodes a digital symbol stream through its modular combination with the delayed result of the modular combination, exemplified by the symbol precoding used in the HDTV transmitters in the The above technique is defined as the "recoding of symbols of the second type" in this specification. The problem of co-channel interference from analog television signals can be observed from the point of view of being a The problem sometimes involves interference in the receiver, to be solved by the adaptive filter circuitry in the receiver. As long as the dynamic range of the system channel is not exceeded, so interference Since the co-channel can capture the channel of the system by destroying the signal transmission capacity for the DTV type modulation, the operation of the system can be observed as an overlap of signal problems. Set field circuits in the receiver is adapted * to select the digital signal from the co-channel interference caused by the analog television signals, which rely on the pronounced correlation and on the anti-correlation properties of the analog television signals to reduce their energy sufficiently to capture the channel system from these. As long as the co-channel interference is related to the signals of analog television, this one introduces the system channel after the TV transmitter and before the DTV transmitter and the DTV receiver. The use or non-use of the precoding of symbols in the DTV transmitter has no effect on the co-channel interference from analog television signals. In the DTV receiver, as long as the co-channel interference is not so large as to overlap the front end of the receiver and capture the channel of the system, it is It is advantageous to precede the data section circuitry with a comb filter to reduce the energy of the higher energy spectral components of the co-channel interference, thereby reducing the errors occurring during the data sectioning. The DTV diffuser must * Adjust your carrier frequency, which is nominally 310 KHz above the. lower limit frequency of the television channel allocation, so that its carrier frequency is optimally shifted in frequency of the video carrier of an analog NTSC co-channel TV signal, which is likely to interfere. This optimal displacement in the carrier frequency is * exactly 59.75 times the frequency of the line fri horizontal scan of the NTSC analog TV signal. The artifacts of the co-channel interference in the demodulated DTV signal will then include oscillations at 59.75 times the frequency of the horizontal scanning line fh of the TV signal NTSC analogue, generated by heterodyning between the digital HDTV bearer and the video carrier of the co-channel analog TV signal, and oscillations at 287.25 times fH, generated by the heterodyne between the digital HDTV carrier and the Chrominance subcarrier of the co-channel analog TV signal, whose oscillations are very close in frequency to the fifth harmonic of the oscillations at 59.75 times fH. The artifacts will also include oscillations of approximately 345.75 times fJ, generated by the heterodynia between * the digital HDTV carrier and the audio carrier of the co-channel analog TV signal, whose oscillations are very close in frequency to the sixth harmonic of the oscillations at 59.75 times f ". The close harmonic relationship of these oscillations allows them all to be suppressed by a properly designed, simple comb filter, which incorporates only a few epochs of differential delay symbols. The use of a filter NTSC type rejection comb before data sectioning in the DTV receiver, incidentally carries out the recoding of symbols of the first type, to modify the symbols obtained by the data sectioning. The data sectioning operation following this recoding of first type symbols in the DTV receiver is a quantization process that is not destructive of the symbols resulting from the recoding of symbols of the first type, as long as the data transmission is compromised, since the data quantization levels are designed to fit the symbol levels. The quantization discriminates against the analog TV signal of co-channel interference, which remains after the filtration associated with the $ recoding of symbols of the first type, and which are appreciably smaller than the steps between the symbol code levels, nonetheless. This is a kind of capture phenomenon in whose phenomenon a stronger signal gains at the expense of a weaker one in a quantification process. Whenever data transmission is related, the stream of digital data symbols flows through the full length of the data channel. system. When the recoding of symbols of the second type is performed as the precoding of symbols in the DTV transmitter, the additive combination of the different currents of delayed data symbols is performed on a modular basis that does not raise the transmitter energy or increase the average intersymbol distance to further assist in overcoming the interference of the analog TV signal. Rather, the main mechanism to overcome the interference of the analog signal TV is its attenuation against the DTV signal, as provided by the comb filtering in the DTV receiver, causing the TV analog signal remaining in the response of the comb filter to be suppressed by the quantization effects in he data disconnector that immediately follows the comb filter. The order of execution of the re-coding procedures of symbols of the first and second types, has no appreciable effect on the transmission of signals through the channel of the system under such circumstances, since neither the coding system destroys the signal transmission capacity for the stream of symbols. The order of realization of the first and second type symbol recoding procedures has no appreciable effect on the digital receiver's ability to suppress the analog, co-channel, analog TV signal, as long as the second symbol recoding type is not interposed between the recoding of symbols of the first type and the subsequent data sectioning. These insights provide the general rationale upon which the inventor based his apparatus described in US Patent Application Serial No. 08 / 839,691 issued April 15, 1997 and entitled "DIGITAL TELEVISION RECEIVER WITH SET OF ADAPTIVE FILTER CIRCUITS FOR DELETE CO-CHANNEL INTERFERENCE ^^ > NTSC. "The adaptive filter circuitry receives a symbol current of 2N level for symbol decoding that can be accompanied by analog signal artifacts of co-channel interference television, where N is a positive integer number. of adaptive filter circuits the NTSC co-channel interference is detected, and it is determined whether this co-channel interference of NTSC has sufficient energy for introducing non-correctable errors in the data sectioning procedure used to perform symbol decoding directly on the recovered baseband signal by synchronously detecting the signal from DTV of VSB-AM. If it is determined that the NTSC co-channel interference does not have sufficient power to cause non-correctable errors, the baseband signal is decoded by symbols using a first data switch to generate results decoding symbols. If it is determined that the NTSC co-channel interference has sufficient power to cause non-correctable errors, the baseband signal is filtered by a first comb-type filter, to reduce the energy of the co-channel interference before being decoded by ^ symbols using a second data disconnector. The first comb filter incidentally carries out a method of recoding symbols of the first type, which introduces error in the symbol decoding results, generated by the second data disconnector. This method of recoding symbols of the first type, carried out before sectioning • Data by the second data disconnector, is observed as a precoding procedure as long as the adaptive filtering is related to suppress NTSC co-channel interference. A second comb filter performs a symbol recoding procedure of the second type, after the data sectioning by the second data disconnector, implementing a post-coding procedure to compensate the re-encoding procedure of symbols of the first type, and generate decoding results of symbols. The symbol recoding procedure of the first type recodes a complete differential delay input symbol stream and the first linear combination of the terms di ferencialmente delayed. The procedure of # recoding of symbols of the second type recodes the decoding results of partially filtered symbols, recovered by the second data disconnector. This method of recoding of symbols of the second type uses a second linear combination of the results of the decoding of partially filtered symbols, with the previous results of the decoding of • symbols fed again with the delay similar to the differential delay introduced in the stream of input symbols, whose second linear combination is checked according to a 2N module arithmetic to generate symbol decoding results, post-coded. The operating errors in the decoded results of decoded symbols are abbreviated or restricted by forcing the symbol decoding results to conform to the ideal results of symbol decoding extracted from the memory in the DTV receiver during the points at which the data field synchronization information occurs and the data segment synchronization information. One of the first and second linear combinations is subtractive, and the other is additive.

The invention relates to the determination of when it is most likely that the symbol decoding results, final, are correct based on the selected estimates of the symbol decoding results, post-encoded, obtained after the comb filtering to suppress the NTSC co-channel interference, rather than the selected estimates of • interim or provisional results of the decoding of symbols, obtained by sectioning data from the baseband symbol code, without having been comb filtered to suppress NTSC co-channel interference. This determination is made by comparing each symbol decoding result, post-coded, to a symbol decoding result, interim, corresponding, over the entire length of each data segment. The obtaining or substantial output of a decoding result of The symbol, post-coding, resulting from the corresponding interim symbol decoding result, is presumed to arise due to the presence of NTSC artifacts in the baseband symbol code, and thus the decoding results of symbols, post-coded, are selected in - ^ _ preference to interim decoding results for inclusion in the final decoding results of symbols, unless other information indicates that such selection could probably be erroneous.

BRIEF DESCRIPTION OF THE INVENTION A receiver of digital signals from The television includes the apparatus for detecting digital television signals, for supplying a stream of 2N level symbols each having a period of symbols of a specified length of time, where N is a positive integer, whose 2N level symbol stream is capable of being accompanied by artifacts of the co-channel analogue television signal. The symbols are grouped into successive data segments with respective headers of the code synchronization of data segments, and the data segments are grouped into successive data fields with the initial data segment of each data field, which contains a data field synchronization code, which changes data field to data field. The digital signal receiver of television includes the set of circuits to provide a number N of unique comb filter responses to the 2N level symbol stream, with each comb-5 filter response being less susceptible to being accompanied by artifacts of the analog television signal, of co-channel interference, than the symbol current level 2N. fc A plurality of symbol decoders are included in the digital signal receiver of television for the generation of results of the symbol decoding, estimated, respective. A first of this plurality of symbol decoders generates first decoding results of symbols, estimated, directly in response to the stream of 2N level symbols. Each of the plurality of other symbol decoders generates respective symbol decoding results, estimated, in response to a respective one of the M's. comb filter responses, unique, whose symbol decoding results, estimated, respectively, are post-coded to complete a respective compared filtering for the respective one of the M comb filter responses, unique to from which those results are obtained fc decoding symbols, estimates, respective. In addition to the first symbol decoder, the plurality of symbol detectors includes at least one second symbol decoder for generating second, estimated symbol decoding results. According to the invention, the digital television receiver of k television includes the set of circuits to detect whether or not there is currently output between the first and second symbol decoding results, estimated, and includes the set of circuits of best estimated selection to select the best estimates from the symbol decoding results, estimated, respectively, to generate final symbol decoding results at times between those times between when the synchronization codes occur. The selection of the best estimates by the set of selection circuits of the best estimate, depends on the current output from the first results of the decoding, estimates of each of the other symbol decoding results, estimated. In one embodiment of the invention M is two, the plurality of decoders of symbols that further includes the first symbol decoder, only a second symbol decoder for generating the second estimated symbol decoding results. The set of circuits for selecting the best estimate takes the following form in this embodiment of the invention. A multiplexer is connected to provide the selection capability between the first and second results of > decoding of symbols, estimated for the generation of symbol decoding results, end times between those times when synchronization codes occur. There is a square for the outputs between the first and second results of the decoding of symbols, estimated, to develop squared results as an absolute measure of those outputs, and a mean averager to generate the average of said squared results. A threshold reflector responds to the average of the results of squared exceeding a prescribed threshold value for the conditioning of the multiplexer, to select the second symbol decoding results, estimated, to generate the final decoding results of symbols, to the times between those times when synchronization codes occur, and otherwise the conditions of the multiplexer to select the first symbol decoding results, estimated, to generate the decoding results of symbols, final, at times between those times when the chronization codes occur.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic block diagram of a television digital signal receiver, which uses an NTSC reject comb filter before the decoding of symbols, and a post-deco comb filter after symbol decoding and, according to the invention, using a co-channel interference detector which compares the decoding results of symbols, obtained without taking measures to suppress the NTSC co-channel interference with the decoding results of symbols obtained by taking measurements to suppress NTSC co-channel interference. Figure 2 is a schematic block diagram showing the details of a detector of # co-channel interference, from NTSC for use in the digital television signal receiver of Figure 1, in which the NTSC co-channel interference detector according to one aspect of the invention provides a set type of best estimate selection circuits, to select the best estimates from the various symbol coding results, for * generate symbol decoding results, endings, to times between those times when synchronization codes occur. Figure 3 is a schematic block diagram showing the details of a portion of the digital television signal receiver of the Figure 1, when the NTSC reject comb filter uses a delay of 12 symbols. Figure 4 is a schematic block diagram showing the details of a portion of the digital television signal receiver of the Figure 1, when the NTSC reject comb filter uses a delay of 6 symbols. Figure 5 is a schematic block diagram showing the details of a portion of the digital television signal receiver of the # Figure 1, when the NTSC reject comb filter uses a delay of 2 video lines. Figure 6 is a block diagram diagram showing the details of a portion of the digital television signal receiver of Figure 1, when the NTSC reject comb filter employs a delay of 262 video lines. Figure 7 is a schematic diagram of * blocks showing the details of a portion of the digital television signal receiver of the Figure 1, when the comb rejection filter of NTSC uses a delay of 2 video frames. Figure 8 is a schematic block diagram showing the details of a portion of the digital television signal receiver of Figure 1 for generating the prescribed symbol decoding results during the data synchronization intervals. Figure 9 is a schematic diagram of blocks showing a digital television signal receiver that uses a plurality of NTSC reject, comb-type filters to perform the parallel decoding of symbols. Figure 10 is a mounting diagram showing how the two can be adjusted together 1A and 10B to form a single figure referred to as Figure 10 in the following detailed description, of which Figure 1Q shows the details of the symbol code selection circuitry that can be used in a digital television signal receiver of the type shown in Figure 9. Figure 10A is a schematic block diagram showing the details of the set of circuits in the digital television signal receiver of Figure 9, to generate prescribed symbol decoding results during the data synchronization intervals. Figure 10B is a schematic diagram of blocks showing the details of the circuitry in the digital television signal receiver of FIG. 9, which includes, in accordance with a further aspect of the invention, an additional type of circuit selection set best estimate to select the best estimates from the various symbol decoding results, to generate the decoding results of symbols, endings, to times between those times when the synchronization coding.

* DETAILED DESCRIPTION At various points in the circuits shown in the figures of the drawings, the compensation delays have to be inserted in order that the sequence of operation is correct, as will be understood by those of experience in the electronic design. Unless there is something out * of the ordinary with respect to a requirement In particular, the compensation delay shall not be explicitly referred to in the following specification. Figure 1 shows a digital television signal receiver used to recover corrected error data, whose data are suitable for recording by means of a digital video cassette recorder or for decoding MPEG-2 and displaying on screen on a television set. The DTV signal receiver of Figure 1 is shown as the receiving broadcast television signals from a receiving antenna 8, but can receive signals from a cable network. The broadcast television signals are supplied as an input signal to the electronic elements 10 "front end". The electronic elements 10 of the "front end" generally include a radio frequency amplifier and first detector for converting the radio frequency television signals to intermediate frequency television signals, supplied as an input signal to an intermediate frequency amplifier chain 12 (IF. ) for DTV type residual sideband signals. The DTV receiver is preferably of the plural conversion type with the IF amplifier chain 12 including an IF amplifier to amplify the DTV signals as they are converted to an ultra-high frequency band by the first detector, a second detector to convert the DTV signals amplified to a very high frequency band, and an additional IF amplifier to amplify the DTV signals and convert them to the VHF band. If the baseband demodulation is performed in the digital regime, the chain 12 of the IF amplifier will further include a third detector for converting the amplified DTV signals to a final intermediate frequency band closer to the baseband. Preferably, a surface acoustic wave (SAW) filter is used in the amplifier of IF for the UHF band, to conform to the channel selection response and reject adjacent channels. This SAW type filter cuts rapidly just beyond 5.38 MHz from the suppressed carrier frequency of the VSB DTV signal and the pilot carrier, which is of similar frequency and fixed amplitude. This SAW type filter consequently rejects most of the frequency-modulated sound bearer of any signal • Analog TV, co-channel interference. The The elimination of the FM sound bearer from any analogue co-channel interference TV signal in an IF amplifier chain 12, prevents artifacts from that carrier from being generated when the final IF signal is detected to recover the baseband symbols and prevents such artifacts from interfering with the data sectioning of those baseband symbols, during symbol decoding. The prevention of such artifacts that interfere with the severing of those symbols of base band during symbol decoding, is better than what can be achieved by relying on comb filtering before data sectioning. The final IF exit signals coming from the 12 string of the IF amplifier, they are supplied to a complex demodulator 14, which demodulates the amplitude modulation DTV signal, of residual sideband in the final intermediate frequency band, to recover a real baseband signal and an imaginary baseband signal. The demodulation can be performed in the digital regime after the analog conversion to I digital of a final intermediate frequency band in the range of a few megacycles, as described for example by C.B. Patel et al. In US Patent No. 5,479.49 issued December 26, 1995 and entitled "DIGITAL VSB DETECTOR WITH PHASE TRACER, FOR INCLUSION IN AN HDTV RECEIVER". Alternatively, the demodulation can be performed in the analog regime, in which case the results are usually subject to conversion from analog to digital to facilitate further processing. The complex modulation is preferably carried out by synchronous demodulation in phase (I) and synchronous demodulation in quadrature phase (Q). The digital results of the above demodulation procedures conventionally have precision of 8 bits or more, and describe 2N level symbols that encode N data bits. Currently, 2N is eight in the case where the DTV signal receiver of Figure 1 receives a broadcast through the .aire via the antenna 12, and is sixteen in the case where the DTV signal receiver of Figure 1 receives cable broadcast. The interest of the invention is with the reception of terrestrial, airborne broadcasts, and Figure 1 does not show the portions of the DTV receiver that provides the # decoding of symbols and decoding of error correction for the transmissions received by cable. The symbol synchronizer and the set of circuits 16 equalizer or compensator receives at least the digitized real samples of the baseband signal in phase (channel I) from the complex demodulator 14; in the DTV receiver of Figure 1 the set of circuits 16 is also shown receiving the digitized imaginary samples of the quadrature phase baseband signal (channel Q). The circuitry 16 includes a digital filter with adjustable weighting coefficients, which compensates for the ghosts and the inclination of the received signal. The symbol synchronizer and the circuitry compensator 16 provide the synchronization of symbols or the "de-rotation" as well as the compensation of amplitude and the elimination of ghosts. The symbol synchronizer and the compensating circuitry in which symbol synchronization is achieved before amplitude compensation is known from US Patent No. 5,479,449. In such designs the demodulator 14 will supply the oversampled demodulator response, which contains real and imaginary baseband signals to the symbol synchronizer and to the compensating circuit set 16. After symbol synchronization, the superimposed data are converted to decimals to extract the channel and baseband signal to the normal proportion of symbols, to reduce the proportion or speed of symbols through the digital filtering used for the compensation of amplitude and the elimination of phantoms. The symbol synchronizer and the circuitry In the case of compensation in which amplitude compensation precedes symbol synchronization, "de-rotation" or "phase tracking" is also known to those skilled in the art of designing digital signal receivers.

Each sample of the output signal of the circuit set is resolved to ten or more bits and is, in effect, a digital description of an analog symbol that shows one of the levels (2N = 8). The output signal of the circuit set 16 is carefully controlled in gain by one of several known methods, so that the ideal step levels for the devices are known. # symbols. A method of gain control, Preferred because the response speed of such gain control is exceptionally fast, it regulates the direct component of the actual baseband signal supplied from the entire demodulator 14 to a normalized level of +1.25. This method of Gain control is generally described in U.S. Patent No. 5,479,449 and is more specifically described by C.B. Patel et al. In U.S. Patent No. 5,573,454 issued June 3, 1997, entitled "GAIN CONTROL RECEIVER AUTOMATIC. RADIO FOR RECEIVING HIGH DEFINITION DIGITAL TELEVISION SIGNALS ", and incorporated by reference herein The output signal from circuitry 16 is supplied as an input signal to the synchronous data detection circuit 18, which retrieves the data field synchronization information F and the data segment synchronization information S from the baseband channel I signal, compensated. Alternatively, the input signal to the synchronous data detection circuitry 18 may be obtained before compensation. The channel I signal samples compensated at a normal symbol rate supplied as the output signal from the set of circuits 16 are applied as the input signal to a comb-type filter 20, of NTSC rejection. The comb-type filter 20 includes a first delay device 201 to generate a pair of currents Differentially delayed 2N level symbol and a first linear combiner 202 to linearly combine the differentially delayed symbol currents, to generate the comb filter response 20. As described in U.S. Patent No. 5,260,793, the first delay device 201 can provide a delay equal to the period of twelve level 2N symbols, and the first linear combiner 202 can be a subtracter. Each sample of the output signal of the filter 20 comb type is resolved to ten or more bits - »* - and is, in effect, a digital description of an analog symbol that shows one of (4N-I) = 15 levels. Synchronizer and symbol compensator circuitry 16 is presumed to be designed to suppress the direct derivative component of its input signal (as expressed in digital samples i), whose bypass or direct bypass component has a standardized level from +1.25, and appears in the real base signal supplied from the complex demodulator 14 due to the detection of the pilot carrier. Accordingly, each output signal sample from the circuitry 16 applied as an input signal to the filter 20 comb type is, in effect, a digital description of an analog symbol that shows one of the following normalized levels: -7, -5, -3, -1, +1, +3, +5 and +7. These symbol levels are referred to as "odd" symbol levels and are detected by an odd-level data switch 22, to generate decoding results of interim or provisional symbols of 000, 001, 010, 011, 100, 101, 110 and 111, respectively. Each sample of the output signal of the filter 20 comb type is, in effect, a description digital of an analog symbol that shows one of the following normalized levels: -14, -12, -10, -8, -6, -4, -2, 0, +2, +4, +6, +8, + 10, +12 and +14. These symbol levels are referred to as "even" symbol levels and are detected by an even-level data switch 24, to generate symbol decoding results, precoded from 001, 010, 011, 100, 101, 110, 111 , • 000, 001, 010, 011, 100, 101, 110 and 111, respectively. The data disconnectors 22 and 24 may be of the so-called "hard decision" type, as presumed to this extent in the description, or may be of the so-called "soft decision" type. used in the implementation of a Viterbi decoding scheme. Arrangements are possible in which the odd-level data switch 22 and the even-level data switch 34 are replaced by a single disconnector, using multiplexer connections to move their place in the circuit and to provide deviation or derivation to modify their division or sectioning intervals, but these arrangements are not preferred due to the complexity of the operation.

Synchronizer and symbol compensator circuitry 16 is presumed in the above description which is designed to suppress the direct derivation component of its input signal (as expressed in digital samples), whose direct derivation component has a standardized level of +1.25 and appears in the actual baseband signal supplied from the complex demodulator 14 due to carrier detection pilot. Alternatively, the synchronizer and symbol compensator circuitry 16 is designed to preserve the direct bypass component of its input signal, which simplifies the design of the compensation filter in the set of circuits 16 to some extent. In this case, the levels of sectioning or division of data in the odd-level data disconnector 22 are shifted to take into account the direct derivation component that accompanies the data steps in your input signal. Provided that the first linear combiner 202 is a subtracter, the circuit set 16 is designed either to suppress or to preserve the direct bypass component of its input signal, which does not consequence with respect to the levels of division or data sectioning in the even level data disconnector. However, if the differential delay provided by the first delay device 201 is chosen so that the first linear combiner 202 is an adder, the levels of sectioning or data splitting in the data switch 24 of the even level must be displaced. for i take into account the duplicate direct derivation component that accompanies the data steps in its signal entry. A comb-type filter 26 is used after the data disconnectors 22 and 24 to generate a post-coding filter response to the response of the precoding filter of the filter. type comb. The comb-type filter 26 includes a multiplexer 261 with 3 inputs, a second linear combiner 262, and a second delay device 263 with a delay equal to that of the first delay device 201 in the comb-type filter 20. The second Linear combiner 262 is a modifier 8 if the first linear combiner 202 is a subtracter and is a modulator 8 of module 8 if the first linear combiner 202 is an adder. The first linear combiner 202 and the second linear combiner 262 can be constructed as read-only memories (ROMs) respective to increase the speed of the linear combination operations sufficiently to support the sample rates or proportions, involved. The output signal from the multiplexer 261 provides the response from the post-encoder comb filter 26, and is delayed by the second delay device 263. The second combiner * Linear 262 combines the decoding results of pre-coded symbols from the even-level data switch 24, with the output signal from the second delay device 263. The output signal of the multiplexer 261 reproduces one of the three input signals applied to the multiplexer 261, as selected in the response to the first, second and third states of a multiplexer control signal, supplied to the multiplexer .261 from a controller 28. The first input access of the multiplexer 261 receives the decoding results of ideal symbols supplied from the memory within the controller 28 during the times when the data field synchronization information F and the segment synchronization information S data from the band I channel signal compensated base, are recovered by the set of circuits 18 for synchronous data detection. The controller 28 supplies the first state of the multiplexer control signal to the multiplexer 261 during these times, conditioning the multiplexer 261 to provide, as the final encoding results that are its output signal, the ideal symbol coding results, supplied from the memory inside the controller 28. The odd-level data disconnector 22 supplies decoding results of interim or interim symbols as its output signal to the second input access of the multiplexer 261. The multiplexer 261 is conditioned by the second state of the multiplexer control signal to reproduce the decoding results of interim or interim symbols, such as the final coding results that are its output signal. The second linear combiner 262 supplies the decoded results of post-coded symbols, such as its output signal to the third input port of the multiplexer 261. The multiplexer 261 is conditioned by the third state of the multiplexer control signal for reproduce the decoding results of symbols, post-encoded, like the final encoding results that are your output signal. The operating errors in the decoded results of post-coded symbols from the post-encoder comb filter 26 are abridged or reduced by feedback of the decoding results of ideal symbols supplied from the memory within the controller 28, during the times when the synchronous data detection circuitry 18 retrieves the data field synchronization information F and the data segment synchronization information S. This is an important aspect of the invention, the which will be described in more detail later in this specification. The output signal from the multiplexer 261 in the post-encoder comb filter 26 comprises the results of End decoding of symbols in groups of 3 parallel bits, assembled by a data assembler 30 for the application to a data interleaver 32. The data interleaver 32 switches the assembled data into parallel streams of data for the application to the circuit set 3. 4 trellis type decoder. The trellis decoder circuitry 34 conventionally uses twelve trellis decoders. The results of the trellis-type decoding are supplied from the circuit set 34 of the trellis decoder to the data de-interleaver circuitry 36 for switching. The byte recognition circuitry 38 % converts the output signal of the interleaver data 36 byte of the Reed-Solomon error correction coding for the application of the Reed-Solomon decoder circuitry 40, which performs the Reed-Solomon type decoding to generate a byte stream with corrected errors, supplied to a data descrambler 42. The data descrambler 42 supplies the reproduced data to the rest of the receiver (not shown). The remainder or remnant of a complete DTV receiver will include a classifier of packet, an audio decoder, an MPEG-2 decoder and so on. The rest of a DTV receiver built into a digital tape recorder / player will include the set of circuits to convert the data to a way for recording or recording.

An NTSC co-channel interference detector 44 supplies to the controller 28 an indication of whether the NTSC co-channel interference is of sufficient force to cause non-correctable error in the division or data sectioning performed by the data disconnector 22. If the detector 44 indicates that the NTSC co-channel interference is not of such strength, the controller 28 will supply the second state of the multiplexer control signal to the multiplexer 261 at times different from those times when the data field synchronization information F and the data segment synchronization information S are retrieved by the synchronous data detection circuitry 18.

This conditions the multiplexer 261 to reproduce as its output signal the results of interim or interim symbol signaling, supplied from the odd-level data disconnector 22. If the detector 44 indicates that the NTSC co-channel interference is of sufficient force to cause non-correctable error in the data division sectioning performed by the data disconnector 22, the controller 28 will supply the third state of the control signal from multiplexer to multiplexer 261, at times different from those times when the data field synchronization information F and the data segment synchronization information S are retrieved by the synchronous data detection circuitry 18. This conditions the multiplexer 261 to reproduce as its output signal the decoding results of symbols, post-encoded, provided as the second linear combination results from the second combiner 262. Figure 2 shows a form of the NTSC co-channel interference detector 44, which can be taken in a modality of the invention. A subtracter 441 differentially combines the decoding results of provisional symbols supplied from the odd-level data disconnector 22 and the decoded results of the post-coded symbols, provided as the second linear combination results __ from the second linear combiner 262. If the amount of the NTSC co-channel interference is negligible, and if the random noise in the baseband I channel signal is negligible, these results of decoding of provisional and postcodes must be similar, so that the difference of the output signal from the subtracter 441 must be low. If the amount of NTSC co-channel interference is appreciable, however, the difference of the output signal from the subtracter 441 will not be generally low, but rather will be frequently high. A measurement of the energy in the differential output signal from the subtracter 441 is developed by quadrature of the differential output signal with a quadrat 442 and determining the average of the response of the quadrature over a prescribed short time interval, with a circuit 443 of elaboration of average. The 442 square can be implemented using the read-only memory (ROM). The averaging circuit 443 can be implemented using a delay line memory to store several successive digital samples and a addition to add the digital samples currently stored in the delay line memory. The short-term arithmetic mean of the energy in the differential output signal from the subtracter 441, as determined by the circuit 443 -of averaging, is supplied to a comparator connected digital to provide a threshold detector 444. The threshold at the threshold detector 444 is sufficiently high so as not to exceed the short-term average of the differences in random noise that accompanies the results of decoding of provisional or interim symbols and the decoding results of post-encoded symbols, applied to the subtractor 441. The * threshold is exceeded if the co-channel interference NTSC is of sufficient force to cause non-correctable error in the division or sectioning of data performed by the data disconnector 22. The threshold detector 444 supplies the controller 28 with the indication of whether the threshold is exceeded or not. 15 In an alternative modality of the square 442 can be replaced by another set of circuits to develop absolute measures of the outputs or games between the first and second symbol decoding results, estimated, such as the absolute value circuit, for example. For the calculation of the speed, such circuit of absolute value can be elaborated using the memory of only reading. Figure 3 is a schematic diagram of blocks showing the details of a portion of the digital television signal receiver of Figure 1, using a sort 120 of the NTSC reject comb type filter 20 and a species 126 of the post-coding comb type filter 26. A subtractor 1202 serves as the first linear combiner in the NTSC reject comb filter 120, and a module additive 1262 serves as the second linear combiner in the post-encoding comb 126 filter. The filter 120 comb type NTSC Reject 10 uses a first delay device 1201 which shows a twelve-symbol times delay, and the post-encoding comb-like filter 126 uses a second delay device 1263 which also shows a delay of eras of twelve symbols. The twelve symbol delay shown by each of the delay devices 1201 and 1263 is close to a cycle delay of the analog TV video carrier artifact at 59.75 times the horizontal scan frequency analog TV fH. The delay of twelve symbols is close to five cycles of the television analog chrominance subcarrier artifact at 287.25 times f. The delay of 12 symbols is close to six cycles of the carrier artifact of analog TV sound at 345.75 times fH.

This is the reason that the differentially combined response of the subtractor 1202 to the audio bearer, the video bearer and the frequencies close to the chrominance subcarrier differentially delayed by the first delay device 1201 tends to have reduced co-channel interference. However, in portions of a video signal in which the edges cross a horizontal scan line, the amount of correlation in the video signal analog television at such distances in the horizontal spatial direction, is very low. A 1261 species of the multiplexer 261 is controlled by a control signal of the multiplexer which is in its second state most of the time when it is determined that there is insufficient co-channel interference from NTSC to cause non-correctable error in the output signal from the data switch 22, and that it is in its third state most of the time when it is determined that there is sufficient NTSC co-channel interference to cause non-correctable error in the output signal from the data disconnector 22. The multiplexer 1261 is conditioned by its control signal which is in its third state for feedback the sum of module 8 results of the adder 1262, as delayed epochs of twelve symbols by the delay device 1263, to the adder 1262, as an addend. This is a modular accumulation procedure in which a simple error propagates as an operating error, with the error recurring each time of twelve symbols. The operating errors in the decoding results of post-encoded symbols from the type 126 filter post-coding comb, are abbreviated or reduced by the multiplexer 1261 which is placed in its first state for four-period symbols at the beginning of each data segment, as well as during the entirety of each data segment containing synchronous field. When the control signal is in its first state, the multiplexer 1261 reproduces as its output signal the decoding results of ideal symbols supplied from the memory in the controller 28. The introduction of the results of ideal decoding of symbols in the output signal of multiplexer 1261 interrupts an operating error. Since there are 4 + 69 (12) symbols per data segment, the ideal decoding results of symbols are shifted to back symbols of four epochs in phase each segment of data, so that the operating error can not persist for more than three data segments. Figure 4 is a schematic block diagram showing the details of a portion of the television digital signal receiver of Figure 1, using a sort 220 of the NTSC reject comb filter 20 and a species 226 of the comb type filter 26 of post-coding. The NTSC reject comb filter 220 uses a first delay device 2201 showing a delay of times of six symbols, and filter post combi 226 uses a second delay device 2263 which also shows a delay of times of six symbols. The delay of six symbols shown by each of the delay devices 2201 and 2263 is close to a delay of 0.5 cycles of the artifact of the analog TV video carrier at 59.75 times the analogue horizontal scanning frequency of television f, close to 2.5 cycles of the artifact of the television analog chrominance subcarrier at 287.25 times fri, and close to 3 cycles of any artifact of the analog TV audio carrier at 345.75 times fh. An adder 2202 serves as the first linear combiner in the filter 220 type comb, NTSC reject, and a subtracter 2262 of the module 8 serves as the second linear combiner in the post-coding comb filter 226. Since the delay shown by the delay devices 2201 and 2263 is shorter than the delay shown by the delay devices 1201 and 1263, although the null near frequencies converted from the analogue carrier television frequencies are narrower band, there is more probability that there is good anticorrelation in the signals additively combined by the additive 2202 that the probability of there being good correlation in the differentially combined signals by the subtracter 1202. The deletion of the sound carrier is more poor in the response of the NTSC reject comb filter 220 than in the response of the NTSC reject comb filter 120. However, if the sound carrier of an analog television signal of co-channel interference has been suppressed by the SAW type filtering or a sound trap in the chain 12 of the IF amplifier, the poor sound rejection of the comb filter 220 is not a problem. The responses of the sync peaks are reduced in duration using the filter 220 type NTSC reject comb of Figure 4, instead of the NTSC reject comb filter 120 of Figure 3, so that there is a substantially reduced tendency to overcome the error correction in trellis type decoding and Reed-Solomon type coding. A 2261 sort of multiplexer 261 is controlled by a multiplexer control signal, which is in its second state most of the time when it is determined that there is insufficient NTSC co-channel interference to cause non-correctable error in the output signal from the data disconnector 22, and which is in its third state most of the time when it is determined that there is sufficient co-channel interference of NTSC to cause error no. correctable in the output signal from the data disconnector 22. The multiplexer 2261 is conditioned by its control signal which is in its third state to feedback the summation results of module 8 'of the addition 2262, as the periods of six symbols delayed by the delay device 263, to the adder 2262, as an addend. This is a modular accumulation procedure in which a simple error propagates as an error of operation, with the recurrent error every time of six symbols. Operating errors in decoding results of post-encoded symbols from the comb filter 226 of post-coding are abbreviated or reduced by the multiplexer 2261 which is placed in its first state for times of four symbols at the beginning of each data segment, as well as during the entirety of each data segment that contains the field synchronism. When this control signal is in its first state, the multiplexer 2261 reproduces as its output signal the results of the decoding of ideal symbols supplied from the memory in the controller 28. The introduction of the ideal decoding results of symbols within The output signal of multiplexer 2261 interrupts an operating error. Since there are 4 + 138 (6) symbols per data segment, the ideal symbol decoding results are shifted backwards times of four symbols in phase each data segment, so that the operating error can not persist for more than two data segments. The probability that a prolonged period of operating error in the post-encoder comb filter 226 will be substantially less than in the filter 126 type of post-coding comb, although the operating error recur more frequently and affects twice as many as the twelve interleaved trellis codes. Figure 5 is a schematic block diagram showing details of a portion of the digital television signal receiver of Figure 4É ~ 1 using a type 320 of the NTSC reject comb filter 20 and a type 326 of the type filter. postcoding comb 26. The NTSC reject comb 320 filter uses a first delay device 3201 which shows a delay of 1368 times symbols, whose delay is substantially equal to the time of the two lines of delay. horizontal scanning of an analog signal of television f, and the filter 326 type comb of post-coding uses a second delay device 3263 which also shows such a delay. The first linear combiner in the comb 320 filter NTSC reject 20 is an additive 3202, and the second linear combiner in the post-encoder comb filter 326 is a subtractor 3262 of module 8. A 3261 sort of multiplexer 261 is controlled by a multiplexer control signal # which is in its second state most of the time when it is determined that there is insufficient NTSC co-channel interference to cause non-correctable error in the output signal from the data disconnector 22, and that it is in its third state the largest part of the time when it is determined that there is sufficient NTSC co-channel interference to cause non-correctable error in the output signal from the data disconnector 22. The The DTV receiver preferably contains the circuitry to detect the change between the alternate scan lines in the NTSC co-channel interference, so that the controller 28 can withstand the supply of the third signal state control of the multiplexer 3261 under such conditions. The multiplexer 3261 is conditioned by its control signal which is in its third state to feedback to the sum of module 8 results. of the adder 3262, as times of 1368 symbols, delayed by the delay device 3263, to the adder 3262 as an addend. That is, a modular accumulation procedure in which a simple error propagates as an error of operation, with the recurrent error every time of 1368 symbols. This duration of the symbol code is longer than the duration for a single block of the Reed-Solomon code, so that a simple operating error is easily corrected during Reed-Solomon type decoding. Operating errors in the decoded results of post-encoded symbols from the post-coding comb filter 326 are reduced or abbreviated by the multiplexer 3261 which is placed in its first state for the entirety of each data segment containing the field synchronization. , as well as for periods of four symbols at the beginning of each data segment. When this control signal is in its first state, the multiplexer 3261 reproduces as its output signal the decoding results of ideal symbols, supplied from the memory in the controller 28. The introduction of the decoding results of ideal symbols in the output signals of the 3261 multiplexer interrupts an operating error. The duration of 16.67 milliseconds of an NTSC video field shows the phase slip versus the duration of 24.19 milliseconds of a DTV data field, so that the DTV data segments containing the Field synchronism eventually explore the full background of the NTSC chart. The 525 lines in the background of the NTSC box each contain times of 684 symbols, for a total of 359,100 times of symbols. Since this is somewhat less than 432 times the epochs of 832 symbols in a DTV data segment containing field synchronism, it may be desired, with reasonable confidence, that the longest duration field operation errors of 432 data will be canceled or deleted by multiplexer 3261 reproducing decoding results of ideal symbols, during the DTV data segments containing field synchronism. There is also phase shift between the data segments, for the groups of T? start code of which the ideal symbol decoding results, and the NTSC video scan lines are available. You can estimate times of 359,100 symbols, which is 89,775 times the times of four symbols in a code start group, which are scanned during 89,775 consecutive data segments. Since there are 313 data segments per DTV data field, it can be desired, with reasonable confidence, that operating errors of longer duration that 287 data fields will be canceled or deleted by multiplexer 3261 reproducing decoding results of ideal symbols during code start groups. The two sources of suppression of operating errors are reasonably independent of one another, so that the operating errors of duration more than ^ L prolonged that two hundred or similar data fields, are very improbable.In addition, if the interference co- 10 NTSC channel drops or decreases at a time when the operating error recur, to condition the multiplexer 3261 for the reproduction of the response of the data disconnector 22 as its output signal, the error can be corrected more in time than what could be otherwise the case. The NTSC reject comb 320 filter of Figure 5 is very good at suppressing demodulation artifacts generated in response to horizontal analog television synchronization pulses, as well as by suppressing many of the demodulation artifacts generated in response to vertical analog TV sync pulses and compensation pulses. These artifacts are the highest energy co-channel interference. Except where there is a change from scan line to line Scanning in the video content of the analog television signal over the period of two scanning lines, the NTSC reject comb 320 filter provides reasonably good 5 suppression of that video content notwithstanding its color. The suppression of the FM audio carrier of the analog television signal is reasonably -j? good, in the case where it has not been suppressed by a tracking reject filter in circuit 16 synchronization and symbol compensation. The artifacts of most analog color TV color synchronizations are suppressed in the NTSC reject comb filter 320 response, too. In addition, filtration provided by the reject comb filter 320 • f of NTSC is "orthogonal" to the interference rejection by NTSC built in trellis type decoding procedures. Figure 6 is a schematic diagram of blocks showing the details of a portion of the digital television signal receiver of Figure 1 using a species 420 of the NTSC reject comb filter 20 and a species 426 of the post-coding comb filter 26. The filter 420 type NTSC rejection comb uses a primer delay device 4201 showing a delay of epochs * of 179,208 symbols, whose delay is substantially equal to the period of 262 horizontal scanning lines of a 5-channel analog signal, and the comb-filter 426 of post-encoding uses a second delay device 4261 that also shows such a delay. A - ^ L adder 4202 serves as the first linear combiner in the NTSC reject comb 420 filter and a subtractor 4262 of module 8 serves as the second linear combiner in the post-coding comb filter 426. A species 4261 of the multiplexer 261 is controlled by a multiplexer control signal which is in its second state most of the time when it is determined that there is insufficient NTSC co-channel interference to cause non-correctable error in the output signal from the data disconnector 22, and that it is in its third state. state most of the time when it is determined that there is sufficient NTSC co-channel interference to cause non-correctable error in the output signal from the data disconnector 22. The DTV receiver preferably contains the set of circuits to detect the change from field to field in the NTSC co-channel interference, so that the controller 28 can withstand the provision of the third state of the multiplexer control signal 4261 under such conditions. The multiplexer 4261 is conditioned by its control signal which is in its third state to feed back to the summing results of the module 8 of the adder 4262, as delayed times of 179,208 symbols by the delay device 4263, to adder 4262 as an addend. This is a modular accumulation procedure in which the simple error propagates as an operating error, with the recurrent error each time of 179,208 symbols. This duration of the code The symbol is longer than the duration for a single block of the Reed-Solomon type code, so that a simple operating error is easily corrected during Reed-Solomon decoding. Operating errors in the The decoding results of post-encoded symbols from the post-coding comb filter 426 are abbreviated or reduced by the multiplexer 4261 which is placed in its first state for the entirety of each data segment. that contains the field synchronism, as well as for the times of four symbols at the beginning of each data segment. When this control signal is in its first state, the multiplexer 4261 reproduces as its output signal the decoding results of ideal symbols, supplied from the memory in the controller 28. The introduction of the decoding results of ideal symbols within the Multiplexer 4261 output signal interrupts an error of operation. The maximum number of data fields required to cancel or clear the operating error in the output signal of the multiplexer 4261 is presumably substantially the same as that required to cancel the operating error. in the output signal of the multiplexer 3261. However, the number of times the error occurs in the period is lower by a factor of 131. The NTSC reject comb filter 420 of FIG. 6 suppresses the most artifacts of demodulation generated in response to the TV analog vertical synchronization pulses and the compensation pulses, as well as the suppression of all the demodulation artifacts generated in response to the horizontal synchronization pulses analog television. These artifacts are the Co-channel interference with the highest energy. Also, the NTSC reject comb filter 420 removes the artifacts that arise from the video content of the analog television signal, which does not change from field to field or from line to line, being freed from the stationary patterns regardless of their frequency spatial -ß horizontal or its color. The artifacts of most color color synchronizations analog TVs are suppressed in the response of the NTSC rejection comb 420 type filter, too. Figure 7 is a schematic block diagram showing the details of a portion of the digital television signal receiver of the L Figure 1, using a kind 520 of the NTSC reject comb 20 filter and a 526 sort of post-coding comb type 26 filter. The filter 520 type NTSC reject comb uses a primer delay device 5201 showing a timestamp of 718,200 symbols, whose delay is substantially equal to the period of two frames or frames of a television analog signal, and the comb 526 post-encoding filter uses a second delay device 5261 showing also such a delay. A subtractor 5202 serves as the first linear combiner in the NTSC reject comb filter 520, and an additive 5262 of module 8 serves as the second linear combiner in the comb 526 post-encoding filter. A 5261 species of the multiplexer 261 is controlled by a multiplexer control signal Jfi that is in its second state most of the time, when it is determined that there is insufficient NTSC co-channel interference to cause non-correctable error in the output signal from the data disconnector 22 and which is in its third state most of the time when it is determined that there is sufficient co-channel interference of NTSC to cause non-correctable error in the signal of t > output from the data disconnector 22. The DTV receiver preferably contains the circuit set to detect the change between the alternating frames in the co-channel interference of NTSC, so that the controller 28 can withstand the supply of the third state of the multiplexer control signal 5261 under such conditions. The multiplexer 5261 is conditioned by its control signal which is in its third state for feedback the sum of module 8 results of the Adder 5262, as times of 718,200 symbols by the delay device 5263, to the adder 5262, as an addend. This is a modular accumulation procedure in which a simple error is propagated as an operating error, with the recurring error each time of 718,200 symbols. This duration of the symbol code is longer than fi: the duration for a simple block of the Reed-Solomon type code, so that an operating error simple is easily corrected during Reed-Solomon type decoding. Operating errors in decoding results of post-coded symbols from the post-coding comb filter 526 are abbreviated or abbreviated. reduced by the multiplexer 5261 that is placed •? F in its first state for the entirety of each data segment containing the field synchronism, as well as for the periods of four symbols at the beginning of each data segment. When is this The control signal is in its first state, the multiplexer .5261 reproduces with its output signal the decoding results of ideal symbols, supplied from the memory in the controller 28. The introduction of the decoding results of ideal symbols in the output signal of the multiplexer 5261 interrupts an operating error. The maximum number of data fields required to cancel or clear the operating error in the output signal of multiplexer 5261 is presumably substantially the same as that required to cancel or clear the operating error in the output signal of multiplexer 3261. However, the number of times the error recurs in this period is lower by a factor of 525. The NTSC reject comb filter 520 of FIG. 7 suppresses all demodulation artifacts generated in response to vertical analog television and video synchronization pulses. the compensation pulses, as well as suppresses also all the demodulation artifacts generated in response to horizontal analog television synchronization pulses. These artifacts are the highest energy co-channel interference.

Also, the NTSC reject comb filter 520 removes the artifacts that arise from the video content of the analog television signal that does not change over two frames, getting rid of such non-stationary patterns regardless of their space frequency or color. The artifacts of all the analog color television color synchronizations are suppressed in the response of the NTSC reject comb filter 520, too. One of experience in the television system design technique will discern other correlation and anti-correlation properties in the analog television signals that can be exploited in the design of the NTSC reject filters of other types than those shown in FIGS. and 7. The use of NTSC reject filters having cascaded NTSC reject filters of the types already described, increases the 2N levels of the baseband signals to the (8N-1) data levels. Such filters may be required to overcome particularly bad co-channel interference problems, despite their drawback of reducing the signal-to-noise ratio for random noise interference with symbol decoding. Figure 8 shows a preferred construction of the multiplexer 261 of Figure 1 in greater detail, together with the circuitry for generating the decoding results of ideal symbols applied to the multiplexer 261. The multiplexer 261 comprises the registers of the fc output buffer of read-only memories (ROM) 46, 48, 50 for selectively reading a bus or collective bus of output 2610 3 bits wide from multiplexer 261. The multiplexer 261 further comprises a three-state compensator 2611 for selectively sending the 3-bit wide output signal of a multiplexer 1612 to the bar | collective output 2610 during the times when decoding results are not generated ideal symbols. The multiplexer 261 responds to the NTSC co-channel interference detector 44 which is a ZERO indicating that the NTSC co-channel interference is not of an amplitude such as to cause non-correctable error in the results of decoding interim or provisional symbols K. supplied by the data disconnector 22 to reproduce the decoding results of interim or interim symbols as the input signal to the three state data compensator 2611.

The multiplexer 261 responds to the NTSC co-channel interference detector 44 which is a ONE indicator that the NTSC co-channel interference is of an amplitude such as to cause non-correctable error in the decoding results of NTSC. provisional symbols, supplied by the fc data disconnector 22 for reproducing the decoding results of. precoded symbols, from the second linear combiner 262 as the input signal to the three state data compensator 2611. The circuit set for generating the decoding results of ideal symbols applied to the collective output bar 2610, comprises the ROMs 46 , 48, 50; a generator 52 of symbol synchronization; an address counter 54 for directing the ROMs 46, 48, 50; the interference reset circuitry 56 for resetting or resetting the counter 54; the address decoders 60, 62, 64 to generate readable signals for ROMs 46, 48, 50; and an ÑOR access 92 for controlling the three-state compensator 2611. The address counter 54 counts the input points received at the symbol decoding rate from the generator 52 of symbol clock, whereby successive respectively descriptive directions of the symbols are generated in a data frame. The appropriate portions of these addresses are applied to the ROMs 46, 48, 50 as their entry addresses. He set of interference adjustment circuits 56 fc resets the counter 54 to the appropriate accounts that respond to the data field synchronization information F and to the data segment synchronization information S retrieved by the synchronous data detection circuitry 18 of Figure 1. It is preferable set the counter 54 of I way a more significant group of bits counts the data segment number per data box and in this way a group of less significant bits counts the number of symbols per data segment. This simplifies the design of the interference reset circuitry 56; reduces the bit widths of the input signal to the address decoders 60, 62, 64; and it facilitates the ROMs 46, 48, 50 which are directed by the partial addresses coming from the counter 54, reducing the bit widths of the ROM address. The ROM 46 stores the decoding results of ideal symbols for a field synchronization segment and is selectively enabled for reading by receiving a UNO from the address decoder. 60. The ROM 46 is directed by the group of bits least significant of the output of the counter 54 that counts the number of symbols per data segment group; and the address decoder 60 responds to the most significant group of bits that counts the number of data segments per data frame. The address decoder 60 generates an ON when and only when the data segment portion of the address supplied by the address counter 54 corresponds to the address of an odd field synchronization segment. The ROM 48 stores the decoding results of ideal symbols for an even field synchronization segment and is selectively enabled for reading by receiving a UNO from the address decoder 62. The ROM 46 is directed by the least significant bit group of the counter output 54 which counts the number of symbols per data segment group; and the address decoder 62 responds to the most significant group of bits that the data segment number counts per data frame. The address decoder 62 generates an ON when and only when the portion of the data segment of the address supplied by the address counter 54 corresponds to the direction of an even field synchronism segment. The ROM 50 stores the decoding results of ideal symbols for the start code group at the beginning of each synchronous segment and is selectively enabled for reading by receiving a UNO from the address decoder 64. The ROM 50 corresponds to the two less significant bits of the output of the counter 54; and the address decoder 64 responds to the group of least significant bits of the counter output 54 that counts the number of symbols per data segment group. The address decoder 64 generates a UNO when and only when the data symbol per portion of the data segment account of the address fc supplied by the address counter 54 corresponds to the partial address of a start code group. 20 The ÑOR access 92 receives the responses from the address decoders 60, 62 and 64 in the respective ones of its three input connections. When the decoding results of ideal symbols are available, one of the address decoders 60, 62 and 64 supplied fc a UNO as its output signal, conditioning the ÑOR access 92 to provide a ZERO response to the three-state data compensator 2611. This conditions the three-state data compensator 2611 to display high source impedances to the bit lines of the collective data bar 2610, so that the signal sent from the multiplexer 2611 will not be evaluated on the collective data bar 2610 of 3-bit width from the. multiplexer 2612. During those portions of the data segments for which the decoding results of ideal symbols are not predictable, none of the address decoders 60, 62 and 64 supplies a UNO as its signal output, conditioning the ÑOR access 92 to provide a UNO response to the three-state data compensator 2611. This conditions the 2611 data compensator of three states to show low source impedances to the bit lines of the collective data bar 2610, so that the signal sent from the multiplexer 2612 will be evaluated on the collective bar 2610 of 3-bit width data. Figure 9 shows a modification of a receiver of digital television signals as fc described below, constructed in accordance with a further aspect of the invention, to operate in parallel a plurality of symbol decoders using respective data switches of equal level, each preceded by a different type of reject comb filter. of NTSC and each one succeeded by a comb filter of post-coding, respectively, to compensate the pre-coding introduced by the comb-type filter of the previous NTSC rejection. An equal level A24 data switch converts the response of an NTSC reject filter A20 from a first type of first decoded symbol decoding results, to the application to a type filter post-coding comb A26 of a first type. An equal level data switch B24 converts the response of an NTSC reject filter B20 of a second type, to the second precoded symbol decoding results for the application to a B26 filter type of post-coding comb of a second type. The equal level C24 data switch converts the response of an NTSC reject filter C20 of a third type to third symbol decoding results precoded, for application to a type filter fc post-coding comb C26 of a third type. The prefixes A, B and C in the identification numbers for the elements of Figure 9 are different integers corresponding to the respective integers 1, 2, 3, 4 and 5, when the receiving portions are used as are shown in any of Figures 3 to 7. The set of symbols decoding selection circuits 66 in Figure 9 formulates a best estimate of the correct symbol decoding for the application to the trellis decoding circuitry 34, selecting from the decoding results of ideal symbols, from the results of Decoding of provisional symbols received from the data disconnector 22 and from the various coding results of post-coded symbols, received from the post-coding comb filters A26, B26 and C26. He The best estimate of the symbol decoding results are used to correct the addition procedures in the post-coding comb filters A26, B26 and C26. The comb filter type A20 rejection of NTSC and the circuitry of the filter tupo comb A26 post-coding filters are advantageously chosen to be of the types such as the NTSC reject comb 520 filter and the 526 post-encoder comb filter circuitry set, of Figure 7. This is so despite a considerable cost in memory , since 718,200 symbols have to be stored in each of the delays of 2 video frames 5201 and 5263. However, the storage in delay 5201 of 2 video frames can be used to perform the shorter delays 4201, 3201, 2201, 1201. Also, the storage in the delay of 2 video frames can be used to perform the shorter delays 4263, 3263, 2263, 1263. 15 The demodulation artifacts high-energy in response to TV analog sync pulses, equalizing or equalizing pulses, and color color synchronizations, are all suppressed when the A20 type filter NTSC rejection combs additively combine alternating video frames. Also, the artifacts that arise from the video content of the analogue TV signal that does not change over two frames, are suppressed, getting rid of the stationary patterns regardless of its spatial frequency or its color. The remaining problem of the deletion of the demodulation artifacts relates mainly to the suppression of those demodulation artifacts that arise from the difference of frame to frame in certain pixel locations within the analogue television signal tracker. These demodulation artifacts can be suppressed by intraframe filtering techniques. The NTSC reject comb filter B20 and the post-decoding comb filter B26 circuitry can be chosen to suppress the remaining demodulation artifacts by relying on the horizontal direction correlation, and the comb comb filter C20. NTSC rejection and the C26 comb filter set of post-coding can be chosen to suppress the remaining demodulation artifacts by relying on the correlation in the vertical direction. Consider how such a design decision can be further implemented. If the sound carrier of a co-channel interference TV analog signal is not suppressed by the SAW filtering or a sound in the IF amplifier chain 12, the NTSC reject comb B20 filter and the post-deco comb B26 filter circuitry are advantageously chosen to be of the types similar to the NTSC reject comb 120 filter and the post-coding comb 126 filter-type circuitry of Figure 3. If the sound carrier of an analog television signal of co-channel interference is suppressed by the SAW filtering or a sound trap in the chain 12 of the IF amplifier, the B20 comb rejection filter of NTSC and the circuit pack of the comb filter B26 of post-coding are advantageously chosen to be of the types similar to the NTSC reject comb filter 220 and the circuit pack of the post-coding comb filter 226 of Figure 4. This is because the anti-correlation between video components separated from each other only by times of six symbols, is usually better than the correlation between the video components of the times of twelve symbols one away from the other. The optimal choice of NTSC reject comb C20 filter and the circuit set of the post-ng C26 comb filter is less fc direct, because the choice must be made (in consideration of the interleaving of fields in the analog interference TV signal) of whether to choose the temporarily closest scan line in the same field or the spatially closest line in the field previous, to be combined with the current scan line in the NTSC reject comb C20 filter. The choice of the temporarily closest scan line in the same The field is generally the best choice, since it is less likely that the jump cuts between fields will ruin the NTSC rejection by the C20 comb filter. With such choice, the NTSC reject comb C20 filter and the filter circuitry type C26 post-ng combs are of types similar to the NTSC reject comb 320 filter and the post-ng comb 326 filter set of Figure 5. With the other choice, the NTSC reject comb C20 filter and the post-deng comb filter C26 circuitry are similar in type to the NTSC reject comb filter 420 and to the post-deng comb filter 426 circuitry of FIG. 6. fc The digital receiver apparatus of Figure 9 is modified in other embodiments of the invention to utilize additional, parallel, data sectioning operations, each carried out by a cascade operation of the respective NTSC reject filter followed by a disconnector of data of equal or even level, respectively, by a respective post-ng comb filter. While two sectioning operations are shown Further, parallel, data in Figure 9, the modifications to use the additional data, parallel, sectioning operations can provide the ability to refine the best estimate of the deng result of correct symbols, even more. In the preferred embodiments of the circuit set fc of FIG. 9, the symbol deng selection circuitry 66 selects the deng results of FIG. ideal symbols as final deng results when these are available. When it is established that the NTSC co-channel interference does in fact obtain differences between the deng results of provisional symbols and the different symbol deng results post-d, are presumably attributable to NTSC co-channel interference. Accordingly, the selection of the final deng results by the set of symbols deng selection circuits 66 when the ideal deng results are not available, may be based on the comparison of the various deng results of post-ded symbols. ficated, one with the other and with provisional deng results. The reason is initially desirable to establish that the NTSC co-channel interference that is indeed obtained is that these differences between the various deng results of symbols may also arise during noise reception conditions when the noise "white" fc is of sufficient level to cause substantially more error in deng results of post-end symbols, than in the provisional deng results. The fact that the NTSC co-channel interference is present can be evaluated during the data segments when the field synchronization information occurs, using the technique described in FIG.

US Patent No. 5,594,496 or techniques Similar fc, using different NTSC co-channel interference rejection filters. Preferably, however, the strength of the NTSC co-channel interference is checked periodically on a continuous basis in real time, so that changes in the level of NTSC co-channel interference due to attenuation can be taken into account. or fading or changes in video content. Figure 9 shows such periodic verification that is provided by measuring the 4.5 MHz intercarrier level in the NTSC co-channel interference, as shown by the inventor in the North American patent application.

Serial No. 08 / 821,945 filed on March 21, 1997 and entitled "USE OF INTERACTIVE SIGNALS TO DETECT NTSC INTERFERENCE IN DIGITAL TV RECEIVERS". The DTV signal, as it is converted to IF by the electronic elements 10 of "extreme The amplifier stages in the IF amplifier chain 68 for the NTSC sound signals correspond to different stages of the amplifier stage 68, which is supplied to an IF amplifier chain of almost parallel type for the NTSC sound signals. similar amplifiers in the chain 12 fc IF amplifier for the DTV signals, having substantially linear gain and having the same automatic gain control as the corresponding amplifier stages in the IF amplifier chain 12. The frequency selectivity of the IF amplifier chain 68 is such as to emphasize the response within ± 250 KHz of the NTSC audio carrier and within ± 250 KHz or similar fc of the NTSC video carrier. The filtering methods for establishing the frequency selectivity of the IF amplifier chain 68, can be carried out by the SAW type filtering in a UHF IF amplifier if the circuit set is used plural conversion receiver. The response of the chain 68 of the IF amplifier is supplied to an intercarrier detector 70 which uses the modulated NTSC video carrier, as an exalted carrier for the heterodination of the NTSC audio bearer. to generate the intermediate frequency signal of intercarrier sound, with a carrier frequency of 4.5 MHz. This intercarrier sound IF signal is amplified by an intermediate frequency amplifier 72, intercarrier sound, whose 4.5 MHz IF amplifier 72 supplies the signal IF amplifier of interporting sound, amplified, to an inter-amplitude detector 74. The response of the amplitude detector 74 is supplied to a threshold detector 76. The threshold in the threshold detector 76 is exceeded if the co-channel interference of the amplifier detector 74 is exceeded. NTSC is of sufficient strength to cause probably error in the division or sectioning of data performed by the data disconnector 22. fc The threshold detector 76 supplies the set of symbol decoding selection circuits 66, the indication of whether the threshold is exceeded or not. If the indication is that the NTSC co-channel interference is not strong enough to probably cause an error in the isolation of the data made by the data disconnector 22, this indication conditions the set of selection circuits 66 of the present invention to select the results of decoding provisional symbols, from the disconnector of data 22 as results of the decoding of final symbols, with the proviso that the decoding results of ideal symbols for the current symbol epoch are not available. The time constant in the amplitude detector 74 of intercarrier should be carefully chosen if optimal performance is sought. Since the decoding errors of isolated symbols are correctable, the elimination of short pulse of the output signal from an intercarrier amplitude detector 74 with a fast time constant, it is likely that it is performed for the generation of a control signal to switch the selection of the decoding results of final symbols from the results of decoding of provisional symbols, to the results of symbol decoding, post-coding. A variety of circuit arrangements are available to derive the intercarrier signal by heterodyning between the audio and video carriers of an analog television signal, of co-channel interference. A number of such arrangements are described in the US Patent Application Serial No. 08 / 821,945. Figures 1 OA and 10B show in some detail the set of circuits included within the set of symbols decoding selection circuits 66, for selecting the decoding results of final symbols. Figure 10 is an assembly or assembly diagram that fc shows how Figures 10A and 10B can be adjusted together to provide a complete schematic block diagram of the symbol decoding selection circuitry 66. The symbol decoding selection circuitry 66 has a collective data bus 78, 3 bit wide, which runs from the bottom of Figure 10A fc to the bottom of Figure 10B, and from here to The beginning of a cascade connection of the data assembler 30, of the data interleaver 32, of the trellis decoder circuitry 34, of the data interleaver 36, of the byte construction circuitry 38, the set of Reed-Solomon decoder circuits 40, and data descrambler 42, similar to that shown in Figure 1. Figure 10A shows the circuit set similar to that shown in Figure 8, to selectively apply the results decoding ideal symbols read from the ROMs 46, 48 and 50 to the collective data bar 78. Figure 10B shows the set of circuits for selecting the best estimate for the selection of the decoding results of final symbols during the time intervals when the decoding results of ideal symbols are not available, ie the time intervals between those in which the synchronization codes of the data segment in the data field are supplied in the DTV signal . The three-state data compensator 2611 and the multiplexer 2612 of Figure 8 are replaced by fc the three-state data compensators 080, A80, B80 and C80 in the best estimate selection circuitry of Figure 10B. The three-state data compensator O80 is conditioned by the response of an AND 082 access which is logical UNO to evaluate the symbol decoding results provisional supplied from the odd-level data switch 22 of Figure 9 on the collective output data bus 78. The three-state data compensator A80 is conditioned by the response of an AND access A82 which is logical ONE for To evaluate the decoding results of post-encoded symbols, supplied from the post-coding comb filter A26 of Figure 9 on the collective bar of. 78 data output. The B80 three-state data compensator is conditioned by the response of an AND B82 access fc which is logical UNO for evaluating the decoding results of post-encoded symbols, supplied from the post-coding comb filter B26 of Figure 9, on the collective bar of output data. The three-state data compensator C80 is conditioned by the response of an AND access C82 which is logical UNO, to evaluate the decoding results of post-encoded symbols, supplied from the post-coding comb filter C26 of Figure 9 on the collective data bar 78 output. The ÑOR access 58 of Figure 10A provides its response to the accesses AND 082, A82, B82 and C82 as respective entries to each one of them. so that the compensators of three states 080, A80, B80 and C80 can be conditioned to show low impedances of the source to the bit lines of the collective data output bar 78 only when the results of the Decoding of ideal symbols are not being evaluated on the collective data bar 78 by any of the output compensators of three states of the ROMs 46, 48 and 50. The output signal from the detector of threshold 76 of Figure 9 is a logical UNO when fc the NTSC co-channel interference is of sufficient force to cause error in the data sectioning performed by the data disconnector 22. The output signal from the threshold detector 76 of Figure 9 is applied as a respective input signal to each of the AND accesses A82, B82 and C82, so that the compensators of three states A80, B80 and C80 can be conditioned to show low impedances of the source to the bit lines of the collective data bus 78, only when the NTSC co-efficient interference is of sufficient force to cause an e in the division or sectioning of data carried out by the data disconnector 22. The output signal coming from the threshold detector 76 of Figure 9 is complemented before being applied as an input signal to the AND access 082, so that the three states compensator 080 can be conditioned to show low impedances of the source to the bit lines of the collective bar 78 of output data, only when the NTSC co-channel interference is of sufficient force to cause an e in the division or sectioning of data made by the disconnector of data 22.

Consider now how the selection is made between the post-decoded symbol decoding results, supplied from the post-coding comb filters A26, B26 and C26 of Figure 9, when the output signal from the threshold detector 76 of the Figure 9 is a logical ONE, indicating that the NTSC co-channel interference is of sufficient strength to cause e in the decoding results of provisional symbols, supplied by the data disconnector 22. The result of decoding of post-coded symbols that comes mainly from the result of decoding provisional symbols, in absolute terms, is presumes that it shows such absolute output due to the comb filtering used to achieve that decoding result of post-encoded symbols, has been more effective in suppressing the NTSC co-channel interference artifact than the comb filtering used to achieve the other decoding results of post-encoded symbols. Consequently, the differences between the result of decoding of provisional symbols supplied by the disconnector of data 22 and the decoding results of post-encoded symbols supplied by the post-encoder comb filters A26, B26 and C26 are determined by the digital subtractors A84, B84 and C84 of Figure 10B. The absolute values of these differences are determined by the absolute value circuits A86, B86 and C86, to determine the outputs or absolute items of the results of I decoding of post-code symbols, supplied from comb-type filters post-coding A26, B26 and C26 from the result of decoding of provisional symbols supplied by the data disconnector 22. The absolute value circuits A86, B86 and C86 can be realized using memories of only reading (ROMs) to achieve the best calculation speed than through the selective addition of bits and the addition of UNO. Even the best calculation speed can be achieved by using the ROMs to carry out the procedures of subtraction and absolute value simultaneously. The set of circuits for selecting the best estimate of Figure 10B includes the digital comparators A88, B88 and C88 as well as the compensators of three states O80, A80, B80 and C80 and in addition to the accesses AND 082, A82, B82 and C82. He fc digital comparator A88 determines whether the absolute output of the provisional symbol decoding result of the decoded symbol decoding result supplied by the comb-like filter A26 of post-encoding, equals or exceeds the absolute output or batch from the decoding result of provisional symbols of the post-decoded symbol decoding result, supplied by the filter type B26 postcoding comb, providing a logical UNO if it does and a logical ZERO if it does not. The digital comparator B88 determines whether the absolute departure or departure from the result of decoding provisional symbols t from the The result of decoding of post-code symbols, supplied by the post-encoding filter 20 B26, exceeds the absolute output from the result of decoding of provisional symbols of the result of decoding of post-encoded symbols, supplied by the post-coding comb filter C26, providing a logical UNO if it does so and a logical ZERO if it does not. A C88 digital comparator determines whether the absolute departure or departure from a decoding result of provisional symbols of the post-decoded symbol decoding result supplied by the post-coding comb filter C26, exceeds the absolute departure or departure of the decoding result of provisional symbols from the post-decoded symbol decoding result supplied by the post-coding comb filter A26, supplying a logical ONE if it does and a logical ZERO if it does not. In order that the AND access response A82 is a logical UNO for the conditioning of the three-state compensator A80 to evaluate the decoding result of post-encoded symbols, supplied by the post-encoding comb filter A26 on the collective bar 78 of output data, the digital comparator A88 must determine that the absolute departure or departure from the decoding result of provisional symbols of the decoded result of post-encoded symbols supplied by the comb filter A26 post-encoding, equals or exceeds the departure or absolute output from of the result of decoding of provisional symbols of the result of Decoding of post-encoded symbols, supplied by the comb filter B26 of post-coding at the same time as the digital comparator C88 determines that the departure or output from the result of decoding of provisional symbols of the result of decoding of post-encoded symbols, supplied by the comb-type filter Postcode coding does not exceed the departure or departure resulting from the decoding result of provisional symbols of the result of decoding of post-encoded symbols supplied by the comb-type filter A26 of post-coding. In order that the response of the access AND B82 is a logical UNO for the conditioning of the B80 three-state compensator, to evaluate the decoding result of the post-encoded symbols supplied by the B26 comb filter of post-coding on the bar collective output 78, the digital comparator B88 must determine that the absolute departure or departure from the decoding result of provisional symbols of the result of decoding of post-encoded symbols, supplied by the comb filter B26 of fc post-coding, exceeds the absolute departure or departure from the decoding result of provisional symbols of the post-decoded symbol decoding result, supplied by the post-coding comb filter C26 at the same time as the digital comparator A88 determines that the absolute output from the result of decoding fc provisional symbols of the result of The decoding of the post-encoded symbols supplied by the post-encoder comb filter A26 does not equal or exceed the output from the decoding result of provisional symbols from the result of decoding of post-code symbols supplied by the B26 comb filter of post-coding. In order that the response of the AND access C82 is a logical UNO for the conditioning of the compensator of three states C80, to evaluate the result of decoding of post-encoded symbols, supplied by the comb filter C26 of post-coding on the collective bus of output data 78, the digital comparator C88 must determine that the absolute output from the decoding result of fc provisional symbols of the decoded postcode decoding result supplied by the post-coding comb filter C26, equal to or exceeds the absolute output from the decoding result of provisional symbols of the post-decoded symbol decoding result, supplied by the comb filter A26 of fc postcoding, at the same time as the comparator digital B88 determines that the output from the decoding result of provisional symbols of the decoded result of decoded post-encoded symbols supplied by the comb filter B26 does not exceed the output from the result of decoding of provisional symbols of the decoded result of symbols after decoding, supplied by the comb filter C26 of post-coding. One and only one of the comparators A88, B88 and C88 (the comparator A88 in Figure 10B) is constructed to provide a logical UNO when its two respective inputs are equal in value, to avoid a condition in which none of the data compensators of three states A80, B80 and C80 can be conditioned to excite the lines of fc bits of the collective bar of output data 78 from the low impedances of the source due to the absolute departures or outputs supplied from the absolute value circuits A86, B86 and C86, all being equal. Someone with experience in the design technique of digital communications receivers and familiar with the above specification and their drawings, will be able to design many embodiments of the invention other than those specifically described. This should be taken into account when constructing the scope of the following broader claims. In the following claims, the word "said" is uses whenever reference is made to an antecedent, and the word "el" (la, los, las) is used for grammatical purposes different from the reference again to an antecedent.

It is noted that in relation to this date, the best method known by the applicant to carry out the aforementioned invention, is the conventional one for the manufacture of the objects to which it refers. 25 Having described the invention as above, property is claimed as contained in the following: fc

Claims (15)

1. A digital television signal receiver, characterized in that it comprises: an apparatus for the detection of digital television signals to supply a stream of 2N level symbols each having an epoch of symbols of a specific length in time, N being a positive integer, said The symbol current of level 2N is capable of being accompanied by co-channel interference television analog signal artifacts, said symbols being grouped in the successive data segments with respective headers of the code. By synchronizing data segments, the data segments are grouped into successive data fields fc with the initial data segment of each data field, which contains a data field synchronization code which changes data field to 20 data field; the set of circuits for the provision of an M number of unique comb filter or comb-type responses to said 2N level symbol current, each comb-filter unique response 25 is less likely to be accompanied by artifacts • of 'the analog signal of co-channel interference television that the current of symbols of level 2N; a plurality of symbol decoders for generating estimated symbol decoding results, respectively, a first of the plurality of symbol decoders generates first decoding results of estimated symbols, which respond to said current of fc 2N level symbols, each other of the plurality of 10 symbol decoders generate decoding results of respective, estimated symbols for a respective one of the unique comb filter responses M, whose estimated symbol decoding results, respectively 15 are post-encoded to complete the respective balanced filtering for a respective one of the M fc unique comb filter responses from which the symbol decoding results, estimated, respectively, are obtained. The other of the plurality of symbol decoders includes a second symbol decoder for generating second symbol decoding results, estimates; the circuitry to detect if 25 there is currently or no departure or departure between • first and second symbol decoding results, estimated; and the set of circuits for selecting the best estimate, to select the best estimates from the symbol decoding results, estimated, respectively, to generate final symbol decoding results at times between those times when synchronization codes occur fc , the selection 10 of the best estimates depends on the outputs from the first symbol decoding results, estimates of other symbol decoding results, estimated.

2. The receiver of digital television signals according to claim 1, further characterized in that it comprises: the set of circuits, trellis-type decoder that responds to the symbol decoding results, endings to generate decoding results of correction of errors, internal; and the set of circuits, Reed-Solomon type decoder that responds to the bytes of the decoding decoding results of internal errors, to generate the decoding results of external error correction.

3. The receiver of digital television signals according to claim 1, characterized in that M is one.

4. The digital television signal receiver according to claim 3, characterized in that the set of circuits for selecting the best estimate comprises: a multiplexer connected to provide the selection capability between the first and second symbol decoding results, estimated to generate the final results of symbol decoding at times between those times when synchronization codes occur; the set of circuits to develop the absolute measurements of the outputs between the first and second decoding results of symbols, estimated; an averager to generate the average or arithmetic mean of the absolute measures; and a threshold detector, which responds to the arithmetic mean of said quadrature results which exceed a prescribed threshold value for conditioning the multiplexer to select the second symbol decoding results, estimated to generate the final decoding results of symbols, at times between those times when the synchronization codes occur, and otherwise conditioning the multiplexer for selecting the first decoding results of symbols, estimated, to generate the final symbol decoding results at times between those times when the synchronization codes occur.

5. The digital television signal receiver according to claim 3, characterized in that the best estimate selection circuit set comprises: a multiplexer connected to provide the selection capability between the first and second symbol decoding results, estimated to generate the final results of symbol decoding, at times between those times when the synchronization codes occur; a square for the outputs between the first and second specification results of »Symbols, estimates, to develop quadrature results as an absolute measure of those outputs; an averager to generate the average or arithmetic mean of the quadrature results; and a threshold detector, which responds to the arithmetic mean of said quadrature results that exceed a prescribed threshold value to condition the multiplexer to select the second symbol decoding results, 10 estimated to generate the final results of symbol decoding, at times between those times when synchronization codes occur, and otherwise conditioning the multiplexer to select the first results of 15 decoding of symbols, estimated, to generate the final decoding results of symbols fc to times between those times when the synchronization codes occur.

6. The television digital signal receiver according to claim 5, characterized in that it further comprises: a set of trellis-type decoder circuits that responds to the final results of 25 symbol decoding to generate results fc decoding of internal error correction; and the Reed-Solomon type decoder set that responds to the bytes of the internal error correction decoding results to generate external error correction decoding results. fc

7. The digital signal receiver of 10 television according to claim 1, characterized in that M is at least two.

8. The receiver of digital television signals according to claim 7, 15 further characterized in that it comprises: the set of circuits to determine whether or not the analog television signal of co-channel interference is currently of sufficient level to give rise to error in the first results of 20 decoding of symbols, estimated, to provide a first indication when determining the analog television signal of co-channel interference that is not of sufficient level to give rise to error in the first 25 decoding results of symbols, estimates, and for the provision of a second indication when the co-channel analog television signal is determined to be of sufficient level to give rise to error in the first estimated symbol decoding results; and, within the set of circuits for selecting the best estimate; and the set of circuits that responds to the first indication to select currently the 10 best estimated only from the first decoding results of symbols, estimated, for the generation of the final symbol decoding results at times between those times when the synchronization codes occur.

9. The receiver of digital television signals according to claim 8, characterized in that the set of circuits for selecting the best estimate further comprises: the set of circuits to determine which of the outputs coming from the first decoding results of estimated symbols of other symbol decoding results, estimated, has the largest absolute value, for 25 generate an indication of which other result of fc symbol decoding, estimates are less likely to be an error caused by said artifacts of the analog television signal of co-channel interference; and the set of circuits that respond to said indication from which other symbol decoding results, estimated, are less likely to be in error due to the fc artifacts of the analog interference television signal 10 co-channel when and only when the second indication is concurrently provided, to select the best estimates from the other estimated symbol decoding results, thus indicating that it is less likely that 15 are in error due to the artifacts of the analog signal of co-channel interference television, to generate the final results of the decoding of symbols, at times between those times when the synchronization codes occur.

10. The receiver of digital television signals according to claim 7, further characterized in that it comprises: the set of circuits to derive the 25 intercarrier signal from the heterodinage between the audio and video carriers of the analog television signal of co-channel interference; the circuitry for detecting when the amplitude of such intercarrier signal exceeds a prescribed level, to provide an indication that said co-channel analog television signal is of sufficient level to give rise to an error in the first results fc decoding symbols, estimates generated 10 for the first symbol decoder, and for otherwise providing an indication that the co-channel analog television signal is of insufficient level to cause error in the first decoding results 15 of symbols, estimated; and, within the set of circuits for selecting the best estimate; and fc the set of circuits which responds to the indication that the analog signal of co-channel interference television is of insufficient level 20 as to give rise to error in the first decoding results of symbols, estimated, to cause that the best estimates currently are selected only from the first results of symbol decoding, 25 estimates.

11. The receiver of digital television signals according to claim 10, characterized in that the set of circuits for selecting the best estimate further comprises: the set of circuits to determine which of the outputs from the first symbol decoding results , estimates of other symbol decoding results, estimated fc, has the highest absolute value, for 10 generate an indication of which of the other symbol decoding results, estimated is less likely to be in error caused by the artifacts of the analog television signal of co-channel interference; and the set of circuits that responds to said indication of which of the other results of fc decoding of symbols, is less likely to be in error due to the artifacts of the analog signal of co-channel interference television, when and only when the second indication is concurrently provided, to select the best estimates from other symbol decoding results, estimated, thus indicated by being less likely to be in 25 error due to the artifacts of the analog signal of co-channel interference television, to generate the final symbol decoding results at times between those times when the synchronization codes occur.

12. A receiver of digital television signals, characterized in that it comprises: the apparatus for the detection of digital television signals, for supplying a current of symbols of level 2N having each one, a time of symbols of a specified length of time, being N a positive integer, said stream of 2N level symbols can be accompanied by artifacts of the analog co-channel interference television signal, said symbols being grouped into successive data segments with respective headings of the segment synchronization code of data, the data segments being grouped in successive data fields' with the initial data segment of each data field, which contains a data field synchronization code which changes from data field to data field; a first symbol decoder that generates first decoding results of fc symbols, estimated, that respond to the stream of 2N level symbols; the set of circuits to provide a plurality of unique comb filter responses to the current of 2N level symbols, each unique comb filter response is less likely to be accompanied by artifacts of the analog co-channel interference television signal , that the stream of 2N level symbols; 10 a respective symbol decoder, responsive to each respective one of a plurality of unique comb filter responses, to generate the respective symbol decoding results, estimated, with the respective results 15 of decoding of estimated symbols that are post-coded to complete the respective balanced filtering for a respective one of the unique comb filter responses from which the respective results of 20 decoding of symbols, estimated; the set of circuits for deriving the intercarrier signal from the heterodinage between the audio and video carriers of the analog television signal of co-channel interference; the set of circuits for detecting when the amplitude of such intercarrier signal exceeds a prescribed level, to provide an indication that said analog television signal of co-channel interference is of sufficient level to give rise to an error in the first results of decoding of symbols, estimates generated by the first symbol decoder, and for fc otherwise providing an indication that the 10 co-channel interference television analog signal is of insufficient level to give rise to error in the first estimated symbol decoding results; the set of circuits to detect the 15 the presence of the synchronization codes in said 2N level symbol stream, to provide fc indications of the presence of synchronization codes in the 2N level symbol stream, and to provide an indication of the absence 20 of synchronization codes in the stream of 2N level symbols; the set of circuits that responds to the presence of synchronization codes in the stream of 2N level symbols, which are detected to generate ideal decoding results of symbols for the synchronization codes; the set of circuits to determine which of the estimated decoding results of symbols different from the first symbol decoding result, estimate currently has the largest absolute output from the first symbol decoding result, fc estimated, to generate indications of which of the 10 other symbol decoding results, estimates are less likely to be in error caused by the artifacts of the analog television signal of co-channel interference; a multiplexer to provide the 15 final results of decoding symbols by reproducing one of a currently selected plurality fc of input signals supplied thereto, said plurality of input signals including the first results of the 20 decoding of symbols, estimates, each of the other symbol decoding results, estimates and the ideal decoding results of symbols; the multiplexer is selectively conditioned to provide the results 25 end of the decoding of symbols by the reproduction of the ideal results of symbol decoding, in response to indications of the presence of synchronization codes in the 2N level symbol stream; the multiplexer is selectively conditioned to provide the final results of symbol decoding by reproducing F of the final results of the decoding of symbols, in response to the concurrent provision of 10 the indication of the absence of synchronization codes in the symbol current of level 2N, and the indication that the analog signal of co-channel interference television is of insufficient level as to give rise to error in the 15 first symbol decoding results, estimated; and the multiplexer is selectively conditioned to provide the final symbol decoding results by reproducing another decoding result of 20 symbols estimated, which is less likely to be in error caused by the artifacts of the analog television signal of co-channel interference, in response to the concurrent provision of the indication of the absence of synchronization codes in the 25 symbol level 2N current, the indication of # that the analog television signal of co-channel interference is of sufficient level to give rise to error in the first decoding results of symbols, estimates, and the indication of which of the other estimated symbol decoding results is less likely that is in error caused by the artifacts of the analog television signal of co-channel interference. fc

The digital television signal receiver according to claim 12, characterized in that it further comprises: the trellis-type decoder circuitry which responds to the final decoding results of symbols to generate internal error correction decoding results; and fc the decoder circuit set of Reed-Solomon that responds to 'the bytes of the decoding results of correction of 20 internal errors for the generation of external error correction decoding results.

14. A receiver of digital television signals, characterized in that it comprises: fc the apparatus for the detection of digital television signals, for supplying a current of symbols of level 2N having each, a time of symbols of a specified length of time, N being a positive integer, said stream of symbols of Level 2N is susceptible to being accompanied by artifacts of the co-channel analogue television signal, these symbols being grouped into segments 10 of successive data with respective headers of the data segment synchronization code, the data segments being grouped in successive data fields with the initial data segment of each data field, which contains a code of 15 data field synchronization which changes from data field to data field; a first symbol decoder that generates first symbol decoding results, estimated, that respond to the current of 20 level 2N symbols; the set of circuits to provide a plurality of unique comb filter responses to the current of 2N level symbols, each unique comb filter response is less 25 likely to be accompanied by artifacts from the fc analog signal from co-channel interference television, than the stream of 2N level symbols; a second symbol decoder responding to the first response of the comb filter to generate the second estimated symbol decoding results, the second symbol decoder includes a first post-encoding comb filter to supply the second fc decoding results of symbols in 10 balanced filter response to the first comb filter response; a third symbol decoder that responds to the second response of the comb filter to generate the third results of 15 decoding of symbols, estimates, the third symbol decoder includes a second post-coding comb filter to supply the third symbol decoding results in balanced filter response to the second 20 comb filter response; a fourth symbol decoder that responds to the third response of the comb filter to generate the fourth symbol decoding results, estimated, the fourth 25 symbol decoder includes a third filter fc type of post-coding comb to supply the fourth results of symbol decoding in balanced filter response to the third comb-type filter response; the set of circuits for deriving the intercarrier signal from the heterodinage between the audio and video carriers of the analog television signal of co-channel interference; fc the set of circuits to detect 10 when the amplitude of such intercarrier signal exceeds a prescribed level, to provide an indication that said analog television signal of co-channel interference is of sufficient level to give rise to an error in the first results 15 for decoding symbols, estimates generated by the first symbol decoder, and for providing otherwise an indication that the co-channel analog television signal is of insufficient level to give rise to 20 error in the first decoding results of symbols, estimated; the circuitry to detect the presence of the synchronization codes in said 2N level symbol stream, to supply 25 indications of the presence of codes of synchronization in the stream of 2N level symbols, and to provide an indication of the absence of synchronization codes in the 2N level symbol stream; the circuitry that responds to the presence of synchronization codes in the stream of 2N level symbols, which are detected to generate ideal decoding results of fc symbols for the synchronization codes; 10 means for determining which of the second, third and fourth estimated symbol decoding results currently has the greatest absolute output from the first symbol decoding result, 15 estimates, to generate indications of which of the second, third and fourth symbol decoding results, estimated is less likely to be in error caused by the artifacts of the television analog signal of 20 co-channel interference; a multiplexer for providing the final symbol decoding results, by reproducing one of the plurality of currently selected input signals, 25 supplied to it, the plurality of signals of fc input includes first symbol decoding results, estimated, second symbol decoding results, estimated third symbol decoding, estimated, and fourth symbol decoding results, estimates and ideal decoding results of symbols; the multiplexer is selectively conditioned to provide the final results of 10 decoding of symbols by reproducing the ideal decoding results of symbols, in response to the indications of the presence of synchronization codes in the 2N level symbol stream, the multiplexer is 15 selectively conditioned to provide the final decoding results of symbols, fc by reproducing the first decoding results of symbols, in response to the concurrent provision of the indication of absence of 20 synchronization codes in the 2N level symbol stream and the indication that the analog co-channel interference television signal is of insufficient level to give rise to error in the first symbol decoding results, 25 estimates; the multiplexer is selectively fc conditioned to provide the final symbol decoding results, by reproducing the second symbol decoding results, in response to the concurrent provision of the indication of the absence of synchronization codes in the 2N level symbol stream, the indication that the analog television signal of co-channel interference is sufficiently high to cause an error in the The first decoding results of estimated symbols, and the indication that the second result of decoding of estimated symbols is less likely to be in error caused by the artifacts of the analogue 15 co-channel interference; the multiplexer is selectively conditioned to provide the final symbol decoding results, by reproducing the third symbol decoding results, in response to the 20 concurrent provision of the indication of the absence of synchronization codes in the symbol current of level 2N, the indication that the analog signal of co-channel interference television is of sufficient level to give rise to error in 25 the first results of the decoding of fc estimated symbols, and the indication that the third estimated symbol decoding result is less likely to be in error, caused by the artifacts of the analog television signal of co-channel interference; and the multiplexer that is selectively conditioned to provide the final results of the symbol decoding, by reproducing the fourth results of decoding of symbols, in response to the 10 concurrent provision of the indication of absence of synchronization codes in the 2N level symbol stream, the indication that the co-channel analog television signal is of sufficient level to give rise to error in the The first results of symbol decoding, estimated, and the indication that the fourth estimated symbol decoding result is less likely to be in error caused by the artifacts of the analogue television signal of 20 co-channel interference.

15. The receiver of digital television signals according to claim 14, characterized in that it also comprises: fc the trellis-type decoder circuitry which responds to the final decoding results of symbols, to generate internal decoding results of error correction; and the Reed-Solomon type decoder circuitry that responds to the bytes of the internal error correction decoding results fc, for the generation of external error correction decoding results. % SUM OF THE INVENTION The decoding results of symbols obtained using comb or comb filtering are compared to suppress NTSC co-channel interference artifacts prior to symbol decoding and decoding results of symbols obtained without using such comb filtering before of decoding 10 of symbols, to determine which decoding results of symbols are to be selected as final results of symbol decoding. The decoding results of symbols obtained using comb filtration 15 to suppress artifacts from NTSC co-channel interference, are selected as the final fc results of symbol decoding, when NTSC co-channel interference occurs. Digital television signals are detected for 20 determine if the 4.5 MHz NTSC intercarrier is obtained, in order to confirm the occurrence or occurrence of the NTSC co-channel interference, substantial.