Coded data generation or conversion – Converter compensation
Reexamination Certificate
2002-08-06
2004-03-16
Tokar, Michael (Department: 2819)
Coded data generation or conversion
Converter compensation
C341S120000
Reexamination Certificate
active
06707402
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention relates to a method in accordance with the preamble to patent claim
1
. A method of this type is generally known.
So-called delta-sigma converters are preferably used for the high-resolution conversion of analog useful signals, especially in the audio range. These converters create a data flow with an extremely high pulse rate in a modulator control circuit, wherein the area content of the digital states of the digital output signal represents the analog information. Thus, the digital information is initially received in the time plane, wherein the achievable resolution of the useful signal represents a function of the pulse frequency. However, the pulse frequency cannot be selected optionally high since the inaccuracies of the signal pulse edges in the form of jitter and edge noise are entered directly into the useful information. By using the presently available technologies, a resolution of 24 bit and a dynamic (signal distance between noise and full scale) of approximately 117 dB (A-weighted) is achieved for delta-sigma converters. However, these values are approximately 15 to 25 dB below the dynamic data, which are required in the area of professional audio signal processing.
The practice of dividing the dynamic range to be mastered into several, preferably two, partial ranges, as shown in
FIG. 1
, is generally known for increasing the dynamic scope during the analog-to-digital conversion.
FIG. 1
shows that the input signal S
1
is transmitted in a first signal path via an adder
2
to an amplifier
3
, in which low-level signals are analog amplified, for example by +30 dB. The amplified signal S
2
is supplied in the first signal path to an analog-to-digital converter
5
. The input signal S
1
furthermore is supplied in a second signal path and without amplification to a second analog-to-digital converter
6
. The two digitized signals S
5
and S
6
from the first signal path and the second signal path are then supplied to a digital signal processor
7
for further signal processing. At the output of this processor, the final digital output signal S
7
is generated by assembling the signals S
5
and S
6
.
With small signal amplitudes, which do not overdrive the amplifier
3
or the analog-to-digital converter
5
in the first signal path, the digital signal S
5
is reduced with the aid of the digital signal processor
7
by the amplification of the amplifier
3
in order to generate the output signal S
7
. As a result, the digital noise from the analog-to-digital converter
5
is reduced by the same amount. The digital signal S
5
that is reduced by the amplification of the amplifier
3
is then switched through as output signal S
7
by the digital signal processor
7
. In contrast, the digital signal S
6
at the analog-to-digital converter
6
output remains turned off by the digital signal processor
7
.
With larger signal amplitudes that can no longer be processed by the first analog-to-digital converter
5
, the digital output signal S
5
of this converter is turned off or blanked out with the digital signal processor
7
. The digital output signal S
6
from the analog-digital converter
6
, which is obtained from the non-amplified input signal S
1
, is used by the digital signal processor
7
as output signal S
7
.
In order to avoid an undesirable overdriving of amplifier
3
, a non-linear network can be connected in the negative feedback path of amplifier
3
, for example in the form of back-to-back diodes D
1
, D
2
. As a result, the amplification of amplifier
3
is reduced for large signal amplitudes, which exceed the modulation range of the analog-to-digital converter
5
and are limited there.
The method explained with the aid of
FIG. 1
is known in expert circles as “gain staging” and has, among other things, the following disadvantages:
The resolution of the useful information in the digital output signal S
7
at the changeover point, during the takeover of the digitized signals S
5
, S
6
from the first or second signal path changes abruptly. The changeover point typically is in the center of the dynamic operating range for the useful signal. This abrupt change in resolution above all affects audio programming signals where sounds with high and very low levels occur simultaneously.
The timely detection of the changeover point from the digitized useful signal that changes, specifically with higher-frequency signal shares, is a problem not least because of the extreme band-width delimitation due to principle of analog-to-digital converters.
The balancing accuracy with respect to amplification and time behavior between the analog and digital signal processing with complementary effect must be extremely high to avoid distortions.
The purposeful overdriving of the amplifier
3
in the first signal path poses high requirements. It should operate completely without delay, which is critical especially for the transition from an overdriven to the normal operational state.
Even if the purposeful overdriving of the amplifier
3
in the first signal path functions completely without problems, errors cannot be avoided at the moment of changeover from the signal S
6
to the signal S
5
. This is due to the fact that the digital decimation filter in the analog-to-digital converter
5
contains erroneous data following an overdriving of the analog-to-digital converter
5
and does not start operating without problems until after a typical throughput time in the range of 0.5 to 1 ms.
Using a method of the aforementioned type, it is therefore the object of the invention to avoid an abrupt change in the signal resolution and in the switching operations for the useful signal. The object furthermore is to achieve a better use of the two signal paths, resulting in a higher or maximum possible signal resolution for larger signal amplitudes and thus an increased dynamic scope while, simultaneously, permitting an uncritical signal processing that is simple and error-free with respect to the digital post processing.
This object is solved with the characterizing features in claim
1
.
SUMMARY OF THE INVENTION
Advantageous embodiments and modifications of the inventive method according to claim
1
follow from the dependent claims.
REFERENCES:
patent: 5451955 (1995-09-01), Sullivan
patent: 6097324 (2000-08-01), Myer et al.
patent: 6271780 (2001-08-01), Gong et al.
patent: 6445330 (2002-09-01), Thomsen et al.
patent: 6473013 (2002-10-01), Velazquez et al.
patent: 6492914 (2002-12-01), Katoh
patent: 6515599 (2003-02-01), Kolsrud et al.
patent: 0 431 214 (1991-06-01), None
patent: 0 475 376 (1992-03-01), None
Georg Neumann GmbH
Kinberg Robert
Nguyen Linh V.
Tokar Michael
Venable LLP
LandOfFree
Method for converting an analog signal to a digital 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 for converting an analog signal to a digital signal, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for converting an analog signal to a digital signal will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3212589