Pulse or digital communications – Systems using alternating or pulsating current – Plural channels for transmission of a single pulse train
Reexamination Certificate
1999-12-15
2003-02-04
Vo, Don N. (Department: 2631)
Pulse or digital communications
Systems using alternating or pulsating current
Plural channels for transmission of a single pulse train
C375S279000
Reexamination Certificate
active
06516036
ABSTRACT:
The present invention relates to an apparatus and concomitant method for signal processing. More particularly, this invention relates to a method and apparatus that determines the desired timing phase to decimate an oversampled input signal, e.g., a QAM signal, to reconstruct the underlying data signal.
BACKGROUND OF THE INVENTION
Power and bandwidth are important resources that are carefully conserved by digital transmission systems through the proper selection of modulation and error correction schemes. Quadrature Amplitude Modulation (QAM) is one form of a multilevel amplitude and phase modulation that is frequently employed in digital communication. QAM modulates a source signal into an output waveform with varying amplitude and phase. The QAM output waveform (QAM signal) can be mapped onto a “constellation diagram” having four quadrants of phasor points. The QAM constellation employs the “I” and “Q” components to signify the in-phase and quadrature components, respectively, where a QAM data word or symbol is represented by both the I and Q components.
Generally, an increase in the number of phasor points (finer constellations) within the QAM constellation will permit a QAM signal to carry more information, but the increase in density of the phasor points creates a disadvantage where the transmitted power is no longer constant. In fact, if the average transmitted signal power is limited, the maximum I and Q values are nearly the same for all the QAM levels, thereby causing the constellation points to be closely spaced as the QAM level increases. Since the distance between phasor points on a QAM constellation generally decreases with additional phasor points, it increases the complexity of distinguishing neighboring phasor points, and translates into a more expensive and complex receiver.
Additionally, it is generally known that a continuous-time signal can be represented by a sequence of its samples that are equally spaced. Namely, the Nyquist theory indicates that at least two samples are necessary per cycle at any frequency (Nyquist rate) in order to analyze it. Therefore, the input signal should be bandlimited to less than half the sampling rate in order to eliminate any frequency component outside the Nyquist limitation.
Thus, a receiver will generally oversample the input signal in order to uniquely reconstruct the underlying data signal. Such oversampled input signal is often then subjected to a conventional two-to-one decimation process, that undersamples the input signal (input data sequence) from two samples per unit time T to one sample per unit time T without discriminating which sample to be selected as the output signal.
In applications where the sample selection issue is not critical, the conventional two-to-one decimator is applicable. However, in some applications, the conventional two-to-one decimator cannot be directly used. Namely, it is very critical in some applications as to which samples are kept and which samples are discarded when the two-to-one decimator is applied to the oversampled input signal.
For example, in QAM demodulation applications, the I and Q symbol sequence, which carries signal information, is embedded in a twice oversampled data sequence. Unless the decimator can selectively determine the correct pair of samples, the data could be incorrectly decimated, thereby resulting in the loss of important information.
Therefore, a need exists in the art for a method and apparatus for determining the correct set of samples to retain in applying a decimation process.
SUMMARY OF THE INVENTION
The present invention is a method and apparatus for determining the correct set of samples to retain in applying a decimation process. Namely, the present invention provides an automatic method of determining the timing phase of the desired samples to decimate the oversampled input signal (data sequence), thereby producing the underlying data signal.
Specifically, an instantaneous power signal is generated for the oversampled input signal. The instantaneous power signal is then decimated using two different timing phases that have the same timing rate. The timing rate of the two different timing phases is suitably selected to be one-half of the timing rate that was applied to sample the input signal. Difference values are then obtained on a sample by sample basis between the two decimated instantaneous power signals, where the difference values are then accumulated in an integrator. The accumulated difference values are compared to two thresholds that dictate and control which timing phase should be used to decimate the oversampled input signal.
The premise of the present invention is that the mean power for the desired samples should be greater than the mean power for the undesired samples. As such, as the sum from the integrator approaches one of the thresholds, the output representative of that threshold will be used to select the proper sampling phase signal. Thus, the present invention can automatically determine the desired timing phase to decimate an oversampled input signal to reconstruct the underlying data signal by evaluating the instantaneous power of the oversampled input signal.
REFERENCES:
patent: 4599732 (1986-07-01), LeFever
patent: 5157694 (1992-10-01), Iwasaki et al.
patent: 6104237 (2000-08-01), Mabuchi
patent: 6121828 (2000-09-01), Sasaki
patent: 6134279 (2000-10-01), Soichi et al.
Lu Cheng-Youn
Perlow Randall Bret
Reed, Jr. Charles
Burke W. J.
Nguyen Dung X.
Sarnoff Corporation
Vo Don N.
LandOfFree
Method and apparatus for decimating an oversampled signal does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for decimating an oversampled signal, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for decimating an oversampled signal will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3173788