Coded data generation or conversion – Analog to or from digital conversion – Analog to digital conversion
Reexamination Certificate
1997-10-30
2002-04-09
JeanPierre, Peguy (Department: 2819)
Coded data generation or conversion
Analog to or from digital conversion
Analog to digital conversion
C341S169000
Reexamination Certificate
active
06369737
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to analog-to-digital conversion and, in particular, to a method and apparatus for converting an analog signal of limited dynamic range to a large dynamic range, floating-point, digital representation.
BACKGROUND OF THE INVENTION
Integrating sensors are used in many applications to measure physical quantities. For example, photodiodes are often used as light sensors to measure the intensity of incident light. In such an application, the photodiode is configured to accumulate (integrate) charge that arises from electron-hole pair formation at the junction of the photodiode as light strikes the junction. The amount of charge that accumulates over a period of time is proportional to the intensity of the light impinging on the junction and, therefore, the accumulated charge can be used to quantify the intensity of the light.
For illustrative purposes, this application will only discuss integrating sensor systems used to measure light intensity. However, the methods and apparatus of both the prior art and the present invention apply equally well to integrating sensor systems in general, including those that are used to measure other physical quantities.
A simplified, typical, prior art integrating sensor system
10
is shown in FIG.
1
. The system comprises a sensor
12
, a reset switch
14
, a sampling switch
16
, a capacitor
18
, an analog processing circuit
20
(such as an amplifier, buffer, or a simple conductor), and an analog-to-digital converter (ADC)
22
. Sensor
12
provides an analog signal in response to an applied stimulus. The analog signal from sensor
12
is applied to a first terminal of reset switch
14
and to a first terminal of sampling switch
16
. Reset switch
14
has a second terminal which connects to a reference voltage (e.g. the positive power supply). Sampling switch
16
has a second terminal which connects to a first terminal of capacitor
18
and also connects to an input terminal of analog processing circuit
20
. Capacitor
18
has a second terminal connected to a reference voltage (e.g. ground), and analog processing circuit
20
has an output terminal is connected to an input terminal of ADC
22
. ADC
22
has an output terminal for producing a digital output that corresponds to the analog signal at its input terminal. Although the system of
FIG. 1
could be used in other applications to measure other physical quantities, let us assume that sensor
12
is a photodetector and that sensor system
10
of FIG.
1
. is used to measure the intensity of light that is incident on sensor
12
. Photodetecting sensor
12
could be a photodiode or any other device that produces electrical charge in response to incident light. The rate of charge production (i.e. the photo-current produced by the photodetector) is proportional to the intensity of the incident light that strikes the sensor.
The system of
FIG. 1
operates as follows: To commence the charge integration process, reset switch
14
closes momentarily to reset the sensor to a known state (e.g. zero accumulated charge or to a reference level such as the positive power supply). Reset switch
14
then opens to allow sensor
12
to begin integrating (i.e. collecting charge). After an integration time T, sampling switch
16
closes, thereby allowing the accumulated charge to pass onto charge collecting capacitor
18
and produce a corresponding analog voltage V(T) at the input terminal of analog processing circuit
20
. This voltage V(T) is proportional to the amount of charge that accumulated in sensor
12
over the integration time T. Analog processing circuit
20
then amplifies, buffers, or otherwise processes this analog voltage to produce a corresponding analog signal S(T) at the input of ADC
22
. ADC
22
then converts this analog signal to a digital representation (typically a binary number) at its output.
Although this approach works reasonably well, the range of light intensities that can be measured is limited by the relatively small dynamic range of the system
10
. The relationship between the light intensity and the charge collected Q(T) can be described by Q(T)=IT, where I is the photocurrent from sensor
12
. I is proportional to the intensity of the incident light. As explained above, this charge Q(T) is dumped onto capacitor
18
at the end of the integration cycle to produce a voltage V(T) according to the equation V(T)=Q(T)/C, and this voltage V(T) is then converted by analog processing circuit
20
to a corresponding signal S(T).
Dynamic range is specified as the ratio of the maximum signal swing that analog processing circuit
20
can produce to the minimum signal level from analog processing circuit
20
that can be meaningfully detected. Thus, to quantify the dynamic range of sensor system
10
, let Q
max
denote the maximum charge capacity of sensor
12
and let S
s
denote the maximum signal swing from analog processing circuit
20
(because S
s
is determined either by the sensor's charge collecting capacity Q
max
or by saturating voltage swings in analog processing circuitry
20
, S
s
does not necessarily correspond to Q
max
). Let Q
n
denote the minimum charge noise signal from sensor
12
, and let S
n
be the minimum noise signal from analog processing circuit
20
. S
n
is due to the sum of dark signal noise, thermal noise, and other noise sources such as shot noise in the MOSFET transistors of analog processing circuit
20
. Finally, let S
min
(corresponding to a charge Q
min
) denote the minimum signal which can be meaningfully detected. S
min
=&agr;S
n
where &agr; is typically a small number (i.e. 1-4). Thus, the dynamic range of system
10
is specified by S
s
/S
min
.
For a given integration time T, since the maximum non-saturating photo-current that can be detected is an I
max
corresponding to S
s
and the minimum photo-current that can be detected is an I
min
corresponding to S
min
, the dynamic range of the system can also be specified by I
max
/I
min
=S
s
/S
min
. If the light intensity is too high, the sensor system saturates at S
s
. The sensor system cannot differentiate any incident light that has a higher intensity than that of the critical light intensity that exactly produces I
max
and S
s
. Conversely, if the light intensity is too low, the sensor's signal is dominated by noise and is effectively lost in S
n
(i.e. S(T)≦S
min
). Although varying the integration time varies both I
max
and I
min
(since Q(T)=∫I dt), the dynamic range remains unchanged. In other words, although varying the integration time T will vary the absolute maximum and minimum light intensities that can be measured, the relative light intensity resolution of the system remains unchanged at S
s
/S
min
.
This limited dynamic range can seriously hamper the performance of a sensor system. For example, consider a two-dimensional array of photo-sensors (i.e. pixels) used to capture an image for a digital camera. The luminance of typical natural and man-made scenes often span
5
orders of magnitude. However, the dynamic range of typical light sensors, such as photodiodes, is much smaller. Consequently, when a scene contains a wide range of light intensities such as bright sunlight and dark shade, the resulting image obtained by the photo-sensor array loses many pictorial details. If the integration time T is minimized such that the brightest portions of the scene do not saturate their corresponding photo-sensors, many pictorial details of the darker areas will be lost. These details will be lost because the small signals produced by the photo-sensors corresponding to the darker areas of the scene will not be sufficient to overcome the noise signal S
n
in those photo-sensors. Conversely, if the integration time T is maximized such that details of the darkest portions of the scene are discernible by their corresponding photo-sensors, the brighter areas of the scene will saturate their corresponding photo-sensors, and the pictorial details of the brighter portions of the scene will b
El Gamal Abbas
Fowler Boyd
Yang David
Jean-Pierre Peguy
Lumen Intellectual Property Services Inc.
The Board of Trustees of the Leland Stanford Junior University
LandOfFree
Method and apparatus for converting a low dynamic range... 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 converting a low dynamic range..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for converting a low dynamic range... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2859850