Pulse or digital communications – Systems using alternating or pulsating current – Plural channels for transmission of a single pulse train
Reexamination Certificate
1999-10-21
2002-10-01
Vo, Don N. (Department: 2631)
Pulse or digital communications
Systems using alternating or pulsating current
Plural channels for transmission of a single pulse train
C375S321000, C329S306000, C332S103000
Reexamination Certificate
active
06459741
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field
This invention relates to Digital Television (DTV) and, in particular, to the use of training sequences designed for N-VSB (N-vestigial sideband modulation) systems, as generally specified in the ATSC (Advanced Television Standards Committee) television standards, within the embodiment of an N-squared QAM (Quadrature Amplitude Modulation) receiver. Additionally, this invention relates to the use of QAM and/or offset (also known as “staggered”) equalization algorithms for the purpose of receiving N-VSB DTV signals, as generally specified in the ATSC television standards.
2. Background Art
Digital television (DTV) signals in the USA are broadcast using the Advanced Television Standards Committee (ATSC) television standard modulation system which is eight level vestigial sideband (8-VSB) modulation with a suppressed carrier signal. Conventional 8-VSB receiver designs complex demodulate the received signal with a pilot tone on zero frequency. Under ideal channel conditions, this allows the data symbols to stream on only one of two complex demodulated channels (known as I-channel only processing).
N-VSB, where N equals 8, is the modulation selected for the ATSC standard for terrestrial broadcast of digital television in many countries, including the United States. N equals 16, or 16-VSB, has been proposed as a standard by the ATSC for wired transmission of digital television.
Currently published 8-VSB receiver designs employ equalization algorithms that operate at baseband on only one of the two channels which result from complex demodulation. Modifications to the N-VSB received signal have been developed to allow use of an N-squared QAM (Quadrature Amplitude Modulation) receiver structure. Use of the training sequences designed for the 8-VSB standard will require modification when used within the 64-QAM structure.
Adaptive equalization filters have been used in an attempt to mitigate the distortion effects of the propagation channel. Filter coefficients are adapted through a variety of mechanisms, but all mechanisms are based on estimating an error. The error estimate is used to adjust the adaptive filter coefficients. A few classes of equalization algorithms are summarized below by summarizing how an error vector is formed.
Decision Directed
The error is formed by forcing a decision on each symbol value, assuming the decision is correct, and forming an error between the decision and the received symbol value.
Training Sequence
The error is formed between the stored training sequence symbol value and the received symbol value.
Blind (Reduced Constellation Algorithm)
The error is formed by forcing a decision on each symbol value, where the decisions are selected from a reduced and possibly modified set of decision values and decision boundaries.
Blind (Property Restoration)
The error is formed between an estimate of a constant property of the distortionless waveform and the property computed from the received waveform.
A multitude of equalization algorithms have been developed for 64 QAM algorithms (see, for example, Richard Gitlan, Data Communication Principles). Improvements upon these algorithms for “staggered” or “offset” modulations have been developed by researchers in the field (see, for example, Jerry C. Tu, “Optimum MMSE Equalization for Staggered Modulation”,
IEEE Comm.
, pp.
1401-1406, 1993).
BRIEF SUMMARY OF THE INVENTION
Training sequences designed for N-VSB systems are used within the embodiment of an N-squared QAM receiver, allowing 8-VSB receivers to be designed using methodologies of 64-QAM receiver design. In particular, a receiver designed using such methodologies converts the received modulation into a signal which can be accepted by circuitry for decoding 64 level quadrature-amplitude modulation (64-QAM) signals. This process provides better signal to noise ratio (SNR) reception than the conventional I-channel only decoding circuitry of most 8-VSB receivers. This process also allows use of training and equalizing algorithms developed for 64-QAM receivers which are superior to equivalent algorithms for 8-VSB receivers. This invention can be generalized to N-VSB conversion into M-QAM where M=N
2
.
The 8-VSB ATSC training sequence can be converted to a two channel QAM sequence by first selecting every other symbol and then inverting every other symbol in each of the QAM symbols. This method can be realized by changing the stored sequence in the receiver memory. A preferred embodiment is an adaptation to the m-sequence generator structure described in detail in a subsequent section.
With a proper QAM training sequence, all QAM or offset QAM modulation equalization algorithms employing a training sequence may be applied in receiving an N-VSB (including the 8-VSB ATSC) digital television signal.
Adaptive equalization algorithms for 8-VSB transmissions implemented within the context of a 64 QAM receiver are superior to present single-channel VSB processing receivers (for example, the DTV Grand Alliance Receiver). Present 64 QAM equalization strategies can be employed when receiving an 8-VSB waveform, given removal of the pilot tone and time offset, except when employing a training sequence. Modifications to the 8-VSB training sequence specification are employed for operation within a 64 QAM receiver design.
REFERENCES:
patent: 5982457 (1999-11-01), Limberg
patent: 6118498 (2000-09-01), Reitmeier
Grabb Mark Lewis
Hershey John Erik
Welles II Kenneth Brakeley
Breedlove Jill M.
Thompson John F.
Vo Don N.
LandOfFree
Implementation of N-VSB training sequences in N-squared QAM... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Implementation of N-VSB training sequences in N-squared QAM..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Implementation of N-VSB training sequences in N-squared QAM... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2980077