US RE42317 E1 Abstract A system for filtering digital television signals is provided. The system comprises a generator for providing a first data sequence to a private data packetizer, and a transmitter for transmitting the packetized first data sequence in a data channel of a digital television signal. The system further includes a receiver for receiving the digital television signal and recovering the first data sequence. The receiver includes a channel estimator for providing an estimate of channel characteristics, such as estimated channel impulse estimate and estimated noise variance. The receiver further includes an adaptive equalizer filter having an input for receiving the digital television signal and an input for receiving adaptive filter coefficients. The receiver further includes a coefficient processor for calculating adaptive filter coefficients based on the channel estimate, and providing the adaptive filter coefficients to the adaptive equalizer filter. The digital television signal is thus filtered to remove undesired channel effects.
Claims(68) 1. A method of filtering a digital television transmission comprising the steps of:
generating a plurality of packetized first data sequencesequences distributed over a training interval at a transmitter;
transmitting through a channel, a digital television stream including said packetized first data sequencesequences;
receiving said digital television stream at a receiver and recovering said first data sequencesequences from said digital television stream;
comparing ones of said first data sequencesequences to a second data sequence, said second data sequence being locally generated, to provide a channel estimate;
applying said received television bit stream to an adaptive filter;
adaptively adjusting filter coefficients of said adaptive filter according to said channel estimate such that undesirable channel effects upon said received television stream are filtered from said received television stream, wherein the digital television transmission is a digital television signal and said data sequences are transmitted in a private channel of an MPEG (Motion Picture Expert Group) data channel.
2. The method of filtering a digital television transmission according to
3. The method of filtering a digital television transmission according to
4. The method of filtering a digital television transmission to according to
5. The method of filtering a digital television transmission according to
generating a packetized first data sequence at a transmitter; transmitting through a channel, a digital television stream including said packetizing first data sequence; receiving said digital television stream at a receiver and recovering said first data sequence from first data sequence from said digital television stream; comparing said first data sequence to a second data sequence, said second data sequence being locally generated, to provide a channel estimate; applying said received television bit stream to an adaptive filter; adaptively adjusting filter coefficients of said adaptive filter according to said channel estimate such that undesirable channel effects upon said received television stream are filtered from said received television stream, wherein said first data sequence is corrupted by noise after passage through said channel to said receiver and is used to compute an estimate of channel frequency response, and wherein the step of comparing comprises the steps of:
computing a Fast Fourier Transform (FFT) of said first data sequence corrupted by noise; computing a FFT of said second data sequence; and dividing the FFT of said first data sequence by the FFT of said second data sequence to provide said estimate of channel frequency response.
6. The method of filtering a digital television transmission according to
7. The method of filtering a digital television transmission according to
8. The method of filtering a digital television transmission according to
9. The method of filtering a digital television transmission according to
convolving said windowed IFFT with said second data sequence to generate an estimated noiseless output;
subtracting said estimated noiseless output from said first data sequence corrupted by noise to generate a difference signal; and
computing an average energy estimation from said difference signal.
10. The method of filtering a digital television transmission according to
11. The method of filtering a digital television transmission according to
12. The method of filtering a digital television transmission according to
generating a packetized first data sequence at a transmitter; receiving said digital television stream at a receiver and recovering said first data sequence from first data sequence from said digital television sequence; applying said received television bit stream to an adaptive filter; adaptively adjusting filter coefficients of said adaptive filter according to said channel estimate such that undesirable channel effects upon said received television stream are filtered from said received television stream, wherein said first data sequence is transmitted in a dynamic or rolling frame/packet structure. 13. A systemAn apparatus for filtering a digital television signal comprising:
a generator for generating first data sequences at a transmitter and the transmitter for broadcasting said digital television signal including said first data sequences in a broadcast channel;
a receiver for receiving the digital television signal, said receiver including:
a channel estimator for comparing said first data sequences to second data sequences, said second data sequences being locally generated, and for providing an estimate of the impulse response of said channel at an output of said channel estimator; and
an adaptive equalizer filter including an input for receiving said digital television signal, and filter taps in communication with said output of said channel estimator such that filter coefficients of said adaptive filter are adjusted according to said estimate of said impulse response of said channel, wherein the digital television signal comprises a high definition television (
HDTV) signal and said first data sequences are transmitted in a private data stream of an MPEG (Motion Picture Expert Group) channel. 14. The apparatus for filtering a digital television signal according to
15. The apparatus for filtering a digital television signal according to
16. The apparatus for filtering a digital television signal according to
17. The apparatus for filtering a digital television transmission according to
a generator for generating first data sequences at a transmitter and the transmitter for broadcasting said digital television signal including said first data sequences in a broadcast channel; a receiver for receiving the digital television signal, said receiving including; a channel estimator for comparing said first data sequences to second data sequences, said second data sequences being locally generated, and for providing an estimate of the impulse response of said channel at an output of said channel estimator; and an adaptive equalizer filter including an input for receiving said digital television signal, and filter taps in communication with said output of said channel estimator such that filter coefficients of said adaptive filter are adjusted according to said estimate of said impulse response of said channel, wherein the digital television signal comprises a high definition television (HDTV) signal and said first data sequences are corrupted by noise after passage through the channel to the receiver, said first data sequences corrupted by noise being used to compute an estimate of the frequency response of said channel, wherein said channel estimator comprises:
a first Fast Fourier Transform (FFT) processor for computing a FFT of said first data sequences corrupted by noise; a generator for generating the second data sequences at the receiver;
a second FFT processor for computing a FFT of said second data sequences; and
a divider for dividing an output of said first FFT processor by an output of said second FFT processor to produce said estimate of channel frequency response.
18. The apparatus for filtering a digital television signal according to
19. The apparatus for filtering said digital television signal according to
20. The apparatus for filtering a digital television signal according to
21. The apparatus for filtering a digital television signal according to
a convolver for convolving said windowed IFFT with said second data sequences to generate an estimated noiseless output;
a subtractor for subtracting said estimated noiseless output from said first data sequences corrupted by noise to generate a difference signal; and
a processor for computing an average energy estimation from said difference signal.
22. The apparatus for filtering a digital television signal according to
23. The apparatus for filtering a digital television signal according to
24. The apparatus for filtering a digital television signal according to
a receiver for receiving the digital television signal, said receiver including: an adaptive equalizer filter including an input for receiving said digital television signal, and filter taps in communication with said output of said channel estimator such that filter coefficients of said adaptive filter are adjusted according to said estimate of said impulse response of said channel, wherein said first data sequences are transmitted in a dynamic or rolling frame/packet structure. 25. A method of processing a digital television stream including a plurality of first data sequences distributed over a training interval transmitted through a channel comprising the steps of:
receiving said digital television stream at a receiver and recovering said first data sequence from said digital television stream; comparing at least one of said first data sequences to a second data sequence, said second data sequence being locally generated; applying said received television stream to an adaptive filter; adaptively adjusting filter coefficients of said adaptive filter according to said comparing step such that undesirable channel effects upon said received television stream are filtered from said received television stream, wherein the digital television stream is a high definition television (HDTV) signal and said first data sequences are transmitted in a channel of an MPEG (Motion Picture Expert Group) data channel. 26. The method of processing a digital television stream according to
27. The method of processing a digital television stream to according to
28. The method processing a digital television stream according to
29. A method of processing a digital television stream including a packetized first data sequence transmitted through a channel comprising the steps of:
applying said received television stream to an adaptive filter; adaptively adjusting filter coefficients of said adaptive filter according to said channel estimate such that undesirable channel effects upon said received television stream are filtered from said received television stream, wherein said first data sequence is corrupted by noise after passage through said channel to said receiver and is used to compute an estimate of channel frequency response, wherein the step of comparing comprises the steps of: computing a Fast Fourier Transform (FFT) of said first data sequence corrupted by noise; computing a FFT of said second data sequence; and dividing the FFT of said first data sequence by the FIT of said second data sequence to provide said estimate of channel frequency response. 30. The method of processing a digital television stream according to
31. The method of processing a digital television stream according to
IFFT) of said quotient of the dividing step. 32. The method of processing a digital television stream according to
33. The method of processing a digital television stream according to
convolving said windowed IFFT with said second data sequence to generate an estimated noiseless output; subtracting said estimated noiseless output from said first data sequence corrupted by noise to generate a difference signal; and computing an average energy estimation from said difference signal. 34. The method of processing a digital television stream according to
35. A method of processing a digital television stream including a packetized first data sequence transmitted through a channel comprising the steps of:
applying said received television stream to an adaptive filter; adaptively adjusting filter coefficients of said adaptive filter according to said channel estimate such that undesirable channel effects upon said received television stream are filtered from said received television stream, wherein said first data sequence is transmitted in a dynamic or rolling frame/packet structure. 36. A method of transmitting a digital television stream through a channel comprising the steps of:
generating a plurality of packetized first data sequences distributed over a training interval to be compared in a receiver coupled to said channel to a second data sequence, said second data sequence being locally generated, to adaptively adjust filter coefficients of an adaptive filter such that undesirable channel effects on a television stream received from said channel are filtered from said received television stream; and transmitting through said channel, a digital television stream including said packetized first data sequences, wherein said first data sequences are transmitted in a private channel of an MPEG (Motion Picture Expert Group) data channel. 37. The method of transmitting a digital television stream according to
38. The method of transmitting a digital television stream to according to
39. The method of transmitting a digital television stream according to
40. A method of transmitting a digital television stream through a channel comprising the steps of:
generating a packetized first data sequence to be compared in a receiver coupled to said channel to a second data sequence, said second data sequence being locally generated, to provide a channel estimate for adaptively adjusting filter coefficients of an adaptive filter such that undesirable channel effects on a television stream received from said channel are filtered from said received television stream; and transmitting through said channel, a digital television stream including said packetized first data sequence, wherein said first data sequence is corrupted by noise after passage through said channel to said receiver, said first data sequence corrupted by noise being used to compute an estimate of channel frequency response, wherein the step of comparing comprises the steps of: computing a Fast Fourier Transform (FFT) of said first data sequence corrupted by noise; computing; a FFT of said second data sequence; and dividing the FFT of said first data sequence by the FFT of said second data sequence to provide said estimate of channel frequency response. 41. The method of transmitting a digital television stream according to
42. The method of transmitting a digital television stream according to
IFFT) of said quotient of the dividing step. 43. The method of transmitting a digital television stream according to
44. The method of transmitting a digital television stream according to
computing an average energy estimation from said difference signal. 45. The method of transmitting a digital television stream according to
46. A method of transmitting a digital television stream through a channel comprising the steps of:
transmitting through said channel, a digital television stream including said packetized first data sequence, wherein said first data sequence is transmitted in a dynamic or rolling frame/packet structure. 47. An apparatus for processing a digital television signal including a plurality of first data sequences distributed over a training interval transmitted in a broadcast channel comprising:
a receiver for receiving said digital television signal, said receiver including; a comparator for comparing said first data sequences to second data sequences, said second data sequences being locally generated; and an adaptive equalizer filter including an input for receiving said digital television signal, and being in communication with said output of said comparator such that filter coefficients of said adaptive filter are adjusted in response to comparator, wherein said first data sequences are transmitted in an MPEG Motion Picture Expert Group) channel. 48. The apparatus for processing a digital television signal according to
49. The apparatus for processing a digital television signal according to
50. The apparatus for processing a digital television signal according to
51. An apparatus for processing a digital television signal including first data sequences transmitted in a broadcast channel comprising:
a receiver for receiving said digital television signal, said receiver including: an adaptive equalizer filter including an input for receiving said digital television signal, and filter taps in communication with said output of said channel estimator such that filter coefficients of said adaptive filter are adjusted according to said estimate of said impulse response of said channel, wherein said first data sequences are corrupted by noise after passage through said channel to said receiver and wherein said first data sequences corrupted by noise are used to compute an estimate of the frequency response of said channel, wherein said channel estimator comprises: a first Fast Fourier Transform (FFT) processor for computing a FFT of said first data sequences corrupted by noise; a generator for generating the second data sequences at the receiver; a second FFT processor for computing a FFT of said second data sequences; and a divider for dividing an output of said first FFT processor by an output of said second FFT processor to produce said estimate of channel frequency response. 52. The apparatus for processing a digital television signal according to
53. The apparatus for processing a digital television signal according to
IFFT) of the quotient of the divider. 54. The apparatus for processing a digital television signal according to
55. The apparatus for processing a digital television signal according to
a convolver for convolving said windowed IFFT with said second data sequences to generate an estimated noiseless output; a subtractor for subtracting said estimated noiseless output from said first data sequences corrupted by noise to generate a difference signal; and a processor for computing average energy estimation from said difference signal. 56. The apparatus for processing a digital television signal according to
57. An apparatus for processing a digital television signal including first data sequences transmitted in a broadcast channel comprising:
a receiver for receiving said digital television signal, said receiver including; an adaptive equalizer filter including an input for receiving said digital television signal, and filter taps in communication with said output of said channel estimator such that filter coefficients of said adaptive filter are adjusted according to said estimate of said impulse response of said channel, wherein said first data sequences are transmitted in a dynamic or rolling frame/packet structure. 58. An apparatus for transmitting a digital television signal through a channel comprising:
a generator for generating a plurality of first data sequences distributed over a training interval to be compared in a receiver to at least one second data sequence, said second data sequence being locally generated to adjust filter coefficients of an adaptive filter; and a transmitter for transmitting a digital television signal including said first data sequences through said channel, wherein said first data sequences are transmitted in an MPEG (Motion Picture Expert Group) channel. 59. The apparatus for transmitting a digital television signal according to
60. The apparatus for transmitting a digital television signal according to
61. The apparatus for transmitting a digital television transmission according to
a first Fast Fourier Transform (FFT) processor for computing a FFT of said first data sequences corrupted by noise; a generator for generating said second data sequences at the receiver; a second FFT processor for computing a FFT of said second data sequences; and 62. The apparatus for transmitting a digital television signal according to
63. The apparatus for transmitting a digital television signal according to
IFFT) of the quotient the divider. 64. The apparatus for transmitting a digital television signal according to
65. The apparatus for transmitting a digital television signal according to
a processor for computing average energy estimation from said difference signal. 66. The apparatus for transmitting a digital television signal according to
67. An apparatus for transmitting a digital television signal through a channel comprising:
a generator for generating first data sequences to be compared in a receiver to second data sequences, said second data sequences being locally generated to provide an estimate of the impulse response of said channel for adjusting filter coefficients of an adaptive filter; and a transmitter for transmitting a digital television signal including said first data sequences through said channel, wherein the digital television signal comprises a high definition television (HDTV) signal and said first data sequences are transmitted in a private data stream of an MPEG (Motion Picture Expert Group) channel, wherein said first data sequences are encrypted. 68. An apparatus for transmitting a digital television signal through a channel comprising:
a transmitter for transmitting a digital television signal including said first data sequences through said channel, wherein said first data sequences are transmitted in a dynamic or rolling frame/packet structure. Description This is a continuation of application Ser. No. This invention relates in digital television and, more particularly, to methods and apparatus for filtering digital television signals to remove multipath and other undesirable effects upon a digital television signal as the signal propagates through a channel. Digital television is an emerging technology that is the subject of much research both in the United States and Japan. Because of the potential advantages of digital television and the many technical problems associated therewith, research into improved systems and methods for transmitting and receiving digital television signals is increasing. One of the most important prevalent problems associated with digital television signals is the problem of multipath effects. The term multipath, as used herein, refers to the propagation of electromagnetic waves along various paths from the digital television transmitter to the digital television receiver. Multipath effects may arise from fixed structures, such as building walls, acting as reflectors in the transmission channel. Moving objects, such as airplanes, may also cause a multipath condition. Even microreflections in cabling can cause multipath conditions. These structures can cause transmission of the television signal to occur along more than one path from the transmitter to the receiver. As a result, the same signal may be received more than once, and at different times by a single, or multiple, receivers. The result of multipath effects in analog television is to create “ghosts” in the displayed television image. In digital television, the effects of multipath can include moderate to severe degradation in the displayed TV picture and sound. Various methods and systems have been designed to address the problem of multipath. See, for example, P. T. Marhiopoulos and M. Sablatash, “Design of a Ghost Canceling Reference Signal for Television Systems in North America,” Proceedings of Canadian Conference on Electrical and Computer Engineering, Vancouver, BC, Canada, Sep. 14-17, 1993, pp. 660-663. The statistics of multipath ghosts have been studied and compiled by, among others, the BTA (Japan's Broadcasting Technology Association). The BTA, and other concerns, designed a “ghost canceling reference” (GCR) transmitted signal to mitigate these multipath effects. The BTA GCR was found to be less than satisfactory in some cases. While homes with outdoor antennas displayed non-varying (stationary) ghosting conditions which could be largely corrected, those homes with indoor antennas experienced changing (dynamic) ghosts. These ghosting conditions were more prevalent where people were moving about the room or other moving objects were in the signal path. The BTA ghost canceller generally was not able to adequately compensate for these conditions. Therefore, a need remains for a system and method for filtering out, or removing, multipath components from digital television signals, and especially for systems and methods for filtering multipath components from a digital television signal when the multipath component arises from moving objects and dynamic conditions in a transmission channel. In an exemplary embodiment of the invention a system for filtering digital television signals comprises a generator for providing a first data sequence to a private data packetizer, and a transmitter for transmitting the packetized first data sequence in a data channel of a digital television signal. The system further includes a receiver for receiving the digital television signal and recovering the first data sequence. The receiver includes a channel estimator for providing an estimate of channel characteristics such as estimated channel impulse response and estimated noise variance. The receiver further includes an adaptive equalizer filter having an input for receiving the digital television signal and an input for receiving adaptive filter coefficients. The receiver further includes a coefficient processor for calculating adaptive filter coefficients and providing the adaptive filter coefficients to the adaptive equalizer filter. The equalizer filter is in communication with the output of the comparing circuit such that filter coefficients of the adaptive filter are adjusted according to the estimate of the impulse response of the data channel. In one embodiment of the present invention, the television transmission is coded according to a Motion Picture Experts Group (MPEG-2) standard. A method of filtering a digital television transmission comprises the steps of generating a first data sequence at a transmitter and periodically inserting the first data sequence into a digital television bit stream to be transmitted. The method further comprises the steps of transmitting the digital television bit stream through a channel to a receiver, receiving the digital television bit stream and extracting the first data sequence from the digital television bit stream. The extracted first data sequence includes channel induced noise. The method further includes the steps of comparing the extracted first data sequence, including channel induced noise, to a second data sequence. The second data sequence is locally generated, that is, the second data sequence is generated at the receiver and does not include channel induced noise. However, in one embodiment of the present invention the second data sequence contains the same data as the first data sequence. The method further includes a step of provide a channel estimate based on the comparison step. The method further includes the steps of applying the received television bit stream to an adaptive filter and adaptively adjusting filter coefficients of the adaptive filter according to the channel estimate such that undesirable channel effects, such as noise, upon said received television bit stream are filtered from said received television bit stream. The foregoing and other objects, aspects and advantages will be better understood from the following detailed description of a preferred embodiment of the invention with reference to the drawings, in which: The present invention relies upon use of the private data stream of a packetized digital television signal. According to the present invention, transmitter The packetized data sequence is then combined by multiplexer The digital television signal, including the packetized video, audio and private data sequences, is received and pre-processed by receiver front end Private data stream Coefficient processor According to one embodiment of the present invention (not shown), a Least Mean Squares (LMS) algorithm is employed to obtain filter coefficients to be applied to equalizer An embodiment of the invention that avoids this autocorrelation matrix problem is illustrated in FIG. One embodiment of channel estimator The channel impulse response signal An estimate of the noise variance, and hence the channel signal-to-noise ratio (SNR) is determined by computing the average energy of the channel estimation error sequence. The channel estimation error sequence is equal to the difference output of subtractor Once the channel impulse response and channel SNR estimates are available, they are provided to coefficient processor The equalizer structure employed in one embodiment of the invention is the minimum mean square error decision-feedback equalizer (MMSE-DFE) shown in FIG. In one embodiment of the present invention coefficient processor A flow chart of the steps of a method and algorithm of the present invention implemented by coefficient processor This triangular factorization contains all the information needed to compute the optimum MMSE-DFE filter settings and determine the optimum delay parameter Δ. More specifically, the optimum delay, Δ The optimum feedback filter coefficients According to another embodiment of the invention, the data sequence is encrypted. An encryption feature, of any type generally available and known, is added to the data sequence so that the known data sequence will be available only to qualified receivers such as, for example, those subscribers who have paid a periodic of pay-for-view access charge. When applied in combination with other embodiments of the invention, a private data channel packet is received and if it is encrypted, the packet is first decrypted by a receiver. In one embodiment of the present invention this is accomplished by utilizing a keying variable. The un-encrypted or decrypted data bits of the private data channel packet may be used for different purposes, such to provide error detection for other received packets' data, to provide error correction for other received packets' data, to provide data that could be used to remove measured data distortion or distortion that was purposely introduced as part of a pay for quality service, as well as to provide a channel estimate for reduction of multipath effects. Encryption of the data sequence may be added so that it may be available only to qualified receivers. Encryption/decryption is accomplished by a “classical” or “one-key” cryptographic system such as the Data Encryption Standard (DES) as defined in the Federal Information Processing Standard 46 (1977) and in a mode defined in Federal Information Processing Standard 81 (1980). Further cryptographic architectural information regarding the DES and its modes of operation is contained in the article “Data Encryption Standard,” by Hershey and Pomper, and is found in Vol. 5, pp. 227-251 of the Froehlich/Kent Encyclopedia of Telecommunications, Marcel Dekker, Inc. In practice, only digital data to be transmitted is encrypted. As shown in In one embodiment of the invention, a dynamic or rolling frame/packet structure is used as depicted in FIG. The training sequence and staggered slots for countering multipath depends upon there being a large number of different inter-interval spacings. By sending known wave-forms (modulated bits) in training intervals (i.e., the Ts), it is possible to locate, within an interval of time, the delays of the strong multipath components. For example, if two identical transmitted waveform Ts are separated by four time intervals and the corresponding received Ts are not identical, an average value of their difference can be formed, and this average value will reflect the average multipath component at a delay of four time intervals. In the time domain, let x(t) be a known training sequence and assume that x(t+4) was also transmitted. Assume that there is a strong multipath component with delay τ+4. Assume y(t) and y(t+4), respectively, are received. The average difference is computed as δ(t)=y(t+4)−y(t), where this average difference removes the effects of uncorrelated multipath from other intervals. What δ(t) produces is −kx(t−τ) where k is the strengths of the multipath, k<1 and τ While the invention has been described in terms of a single preferred embodiment, those skilled in the art will recognize that the invention can be practiced with modification within the spirit and scope of the appended claims. Patent Citations
Non-Patent Citations
Referenced by
Classifications
Rotate |