Method and system for correcting quantization loss during...

Details Method and system for correcting quantization loss during... Method and system for correcting quantization loss during...

2002-12-12

2003-10-14

Tokar, Michael (Department: 2819)

Coded data generation or conversion

Analog to digital conversion followed by digital to analog...

C341S118000, C341S120000, C341S144000, C341S155000

Reexamination Certificate

active

06633245

ABSTRACT:

FIELD OF THE INVENTION
The present invention relates to analog to digital to analog signal conversion, and more particularly to correcting quantization loss during such a conversion.
BACKGROUND OF THE INVENTION
FIG. 1
is a block diagram illustrating a conventional analog to digital to analog signal conversion system
10
. As is shown, an analog signal
20
is inputted into an analog to digital converter
12
, where the analog signal is converted into a digital discrete time signal
22
. A digital system
14
utilizes the digital signal
22
in such a manner that is well known to those skilled in the art, e.g., for storage, delay, etc., and is not germane to the present invention. The digital discrete time signal
22
is converted back into a reconstructed analog signal
24
by a digital to analog converter
16
. The resulting reconstructed analog signal
24
can be passed through a low pass filter
18
to filter out quantization noise before the signal
24
is finally outputted as an analog signal
26
.
Ideally, the analog signal out
26
is identical to the analog signal in
20
. Quantization losses, however, inherently occur during the conversion process. In particular, quantization losses occur when the analog signal
24
is being reconstructed from the digital discrete time signal
22
. This quantization loss is known as a “(sin X)/x loss,” and is reflected as a diminished energy amplitude for the signal.
The (sin x)/x loss is a function of how often the digital signal
22
is sampled in order to reconstruct the analog signal in
20
. At lower frequency bandwidths of the digital signal
22
, it is not difficult to select a sampling frequency that allows the digital to analog converter
16
to collect enough data points to accurately reconstruct the analog signal in
20
. Thus, at lower frequency bandwidths, the quantization loss is negligible.
At higher frequency bandwidths, however, the sampling frequency is not high enough to accurately capture the true nature of the signal. For instance, if the sampling frequency is only twice the upper frequency of the high frequency bandwidth, the converter
16
will attempt to reproduce the upper frequency based on only two sampled data points. The converter
16
can only estimate the nature of the signal between the two points, and oftentimes underestimates the amplitude peak of the signal. Consequently, the quantization loss can be significant and the high frequency bandwidth of analog signal out
26
will not be an accurate reproduction of the same bandwidth of the analog signal in
20
.
Accordingly what is needed is a system and method for correcting quantization loss during analog to digital to analog signal conversion. The system and method should provide an acceptable degree of accuracy of correction, but not require excessive processing power. The present invention addresses such a need.
SUMMARY OF THE INVENTION
The present invention provides a method and system for correcting quantization loss of a signal during analog to digital to analog signal conversion, wherein the quantization loss is a function of (sin x)/x. The system and method of the present invention includes utilizing a continuous function polynomial to represent a (sin x)/x function, and applying an inverse function of the continuous function polynomial to the signal to provide a correction for the quantization loss.
By utilizing a continuous function polynomial, such as McLaurin's polynomial, to represent the (sin x)/x function, the correction for the quantization loss can be calculated quickly and easily, thereby expending little processing power. Moreover, a filter that creates an inverse function of the continuous function polynomial leaves a small footprint because it does not require significant hardware resources. If a sampling frequency is at least four times an upper frequency of a desired frequency bandwidth, only the first two members of the polynomial need be used to accurately approximate the quantization loss. Accordingly, under these conditions, the correction is accurately provided by creating an inverse function of only the first two members of the polynomial.


REFERENCES:
patent: 4862168 (1989-08-01), Beard
patent: 6016113 (2000-01-01), Binder
patent: 6016327 (2000-01-01), Kaplan
patent: 2002/0118219 (2002-08-01), Nally et al.

Affiliated with

Also associated with

Rating

Method and system for correcting quantization loss during... 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 system for correcting quantization loss during..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for correcting quantization loss during... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3166601