U.S. patent application number 10/348566 was filed with the patent office on 2004-07-22 for optical digital signal processing system and method.
This patent application is currently assigned to Texas Instruments Incorporated. Invention is credited to So, John Ling Wing.
Application Number | 20040141742 10/348566 |
Document ID | / |
Family ID | 32712579 |
Filed Date | 2004-07-22 |
United States Patent
Application |
20040141742 |
Kind Code |
A1 |
So, John Ling Wing |
July 22, 2004 |
OPTICAL DIGITAL SIGNAL PROCESSING SYSTEM AND METHOD
Abstract
A method for optical digital signal processing, comprises
configuring a plurality of binary mirrors to allow a subset of the
binary mirrors to represent a range of values. The plurality of
binary mirrors comprise a digital micromirror device. Light from a
light source is received at the digital micromirror device. The
intensity of the light is altered to represent one of the values
based, at least in part, on the configuration of the subset of the
binary mirrors. The altered light is transmitted from the digital
micromirror device to a detector array.
Inventors: |
So, John Ling Wing; (Plano,
TX) |
Correspondence
Address: |
TEXAS INSTRUMENTS INCORPORATED
P O BOX 655474, M/S 3999
DALLAS
TX
75265
|
Assignee: |
Texas Instruments
Incorporated
|
Family ID: |
32712579 |
Appl. No.: |
10/348566 |
Filed: |
January 20, 2003 |
Current U.S.
Class: |
398/45 |
Current CPC
Class: |
G02B 26/0841 20130101;
G06E 3/00 20130101 |
Class at
Publication: |
398/045 |
International
Class: |
G02F 001/03; G02F
001/07 |
Claims
What is claimed is:
1. A method for optical digital signal processing, comprising:
configuring a plurality of binary mirrors to allow a subset of the
binary mirrors to represent a range of values, the plurality of
binary mirrors comprising a digital micromirror device; receiving
light from a light source at the digital micromirror device;
altering the intensity of the light to represent one of the values
based, at least in part, on the configuration of the subset of the
binary mirrors; and transmitting the altered light from the digital
micromirror device to a detector array.
2. The method of claim 1, wherein altering the intensity of the
light comprises arranging the subset of binary mirrors in a pattern
to improve detectability by the detector array.
3. The method of claim 2, wherein arranging the subset of binary
mirrors in a pattern to improve detectability by the detector array
comprises: activating a first mirror from the subset to be in an on
position; selecting a second mirror from the subset, the second
mirror being positioned diagonally to the first mirror; and
activating the selected second mirror to be in the on position.
4. The method of claim 3, further comprising: selecting a third
mirror from the subset, the third mirror being substantially
adjacent to the first and second mirrors; and activating the
selected third mirror to be in the on position.
5. The method of claim 4, further comprising: selecting a fourth
mirror from the subset, the fourth mirror being substantially
adjacent to the first and second mirrors; selecting a fifth mirror
from the subset, wherein the fifth mirror being positioned
diagonally to at least one of the selected mirrors; activating the
selected fourth mirror to be in the on position; and activating the
selected fifth mirror to be in the on position.
6. The method of claim 1, wherein the subset of binary mirrors
comprises four mirrors and the range of values is between zero and
three.
7. The method of claim 6, wherein the subset of binary mirrors are
in a 4.times.1 configuration.
8. The method of claim 6, wherein the subset of binary mirrors are
in a 2.times.2 configuration.
9. The method of claim 1, wherein the subset of binary mirrors
comprises sixteen mirrors and the range of values is between zero
and fifteen.
10. The method of claim 1, wherein the subset of the binary mirrors
comprises a first subset of binary mirrors and the method further
comprises: receiving the light at a second subset of binary
mirrors; reconfiguring the plurality of binary mirrors to allow the
second subset of the binary mirrors to represent the range of
values; and altering the intensity of the light to represent one of
the values based, at least in part, on the configuration of the
second subset of the binary mirrors.
11. The method of claim 1, further comprising: generating an input
sequence operable to drive the light source; digitizing the
generated input sequence; and communicating the digital input
sequence to the light source.
12. The method of claim 9, further comprising oversampling the
light from the digital micromirror device to the detector
array.
13. The method of claim 12, wherein oversampling the light from the
digital micromirror device to the detector array comprises
optically linking the sixteen binary mirrors to sixteen
detectors.
14. The method of claim 1, wherein the subset of the binary mirrors
comprises a first subset of binary mirrors and the first subset
represents a first nibble of data and the method further comprises:
configuring the plurality of binary mirrors to allow a second
subset of the binary mirrors to represent the range of values, the
second subset of binary mirrors representing a second nibble of
data; and altering the intensity of the light to represent one of
the values based, at least in part, on the configuration of the
second subset of the binary mirrors.
15. The method of claim 14, further comprising linking the first
subset of the binary mirrors and the second subset of the binary
mirrors to represent a byte of data.
16. The method of claim 15, wherein the first subset of the binary
mirrors is weighted by sixteen to the zero power and the second
subset of the binary mirrors is weighted by sixteen to the first
power.
17. The method of claim 16, further comprising: configuring the
plurality of binary mirrors to allow a third subset of the binary
mirrors to represent the range of values, the third subset of
binary mirrors representing a third nibble of data; altering the
intensity of the light to represent one of the values based, at
least in part, on the configuration of the third subset of the
binary mirrors; and linking the third subset of the binary mirrors
with the first and second subsets of the binary mirrors to
represent a 12-bit word, wherein the third subset of binary mirrors
is weighted by sixteen to the second power.
18. The method of claim 17, wherein the second subset of binary
mirrors is positioned vertically adjacent to the first subset of
binary mirrors and horizontally adjacent to the third subset of
binary mirrors.
19. The method of claim 17, wherein the second subset of binary
mirrors is positioned vertically adjacent to the first subset of
binary mirrors and vertically adjacent to the third subset of
binary mirrors.
20. The method of claim 17, further comprising: configuring the
plurality of binary mirrors to allow a fourth subset of the binary
mirrors to represent the range of values, the fourth subset of
binary mirrors representing a fourth nibble of data; altering the
intensity of the light to represent one of the values based, at
least in part, on the configuration of the fourth subset of the
binary mirrors; and linking the fourth subset of the binary mirrors
with the first, second, and third subsets of the binary mirrors to
represent a 16-bit word, wherein the fourth subset of binary
mirrors is weighted by sixteen to the third power.
21. The method of claim 20, further comprising arranging the
subsets of binary mirrors in a pattern to improve detectability by
the detector array.
22. The method of claim 21, wherein the second subset of binary
mirrors is positioned vertically adjacent to the first subset of
binary mirrors and horizontally adjacent to the fourth subset of
binary mirrors and the third subset of binary mirrors is positioned
horizontally adjacent to the first subset of binary mirrors and
vertically adjacent to the fourth subset of binary mirrors.
23. The method of claim 21, wherein the second subset of binary
mirrors is positioned vertically adjacent to the first subset of
binary mirrors and vertically adjacent to the third subset of
binary mirrors and the third subset of binary mirrors is positioned
vertically adjacent to the second subset of binary mirrors and
vertically adjacent to the fourth subset of binary mirrors.
24. The method of claim 1, wherein configuring a plurality of
binary mirrors to allow a subset of the binary mirrors to represent
a range of values comprises configuring the plurality of binary
mirrors to allow the subset of the binary mirrors to represent a
range of negative values.
25. The method of claim 24, wherein altering the intensity of the
light to represent one of the values comprises arranging the subset
of binary mirrors in a photonegative pattern.
26. The method of claim 24, wherein the representation of the range
of negative values is in twos complement.
27. An optical digital signal processing system, comprising: a
light source; a detector array comprising a plurality of detectors;
a digital micromirror device comprising a plurality of rows and
columns of binary mirrors, the digital micromirror device operable
to receive light from the light source and each mirror is operable
to illuminate one of the detectors in response to being in an on
position; and a digital signal processor coupled to the digital
micromirror device and the light source, the digital signal
processor operable to configure the digital micromirror device to
allow a subset of the binary mirrors to represent a range of values
and generate an input sequence operable to drive the light
source.
28. The optical digital signal processing system of claim 27,
wherein the digital micromirror device is further operable to alter
the intensity of the received light to represent one of the values
based, at least in part, on the configuration of the subset of the
binary mirrors.
29. The optical digital signal processing system of claim 28,
wherein: the subset of the binary mirrors comprises a first subset
of binary mirrors; and in response to receiving the light at a
second subset of binary mirrors: the digital signal processor is
further operable to reconfigure the plurality of binary mirrors to
allow the second subset of the binary mirrors to represent the
range of values; and the digital micromirror device is further
operable to alter the intensity of the light to represent one of
the values based, at least in part, on the configuration of the
second subset of the binary mirrors.
30. The optical digital signal processing system of claim 27,
wherein the digital signal processor comprises a reduced
functionality digital signal processor.
31. The optical digital signal processing system of claim 30
further comprising a digital-to-analog converter coupled between
the digital signal processor and the light source.
32. The optical digital signal processing system of claim 31,
wherein the digital signal processor is operable to digitize the
input sequence and wherein the digital-to-analog converter is
operable to: receive the digitized sequence from the digital signal
processor; convert the digitized sequence into an analog current;
and communicate the converted analog current to the light source,
wherein the current is operable to drive the light source.
33. The optical digital signal processing system of claim 27,
wherein the light source comprises a first light source positioned
horizontally in relation to a first column of the binary mirrors
and the optical digital signal processing system further comprises:
a second light source positioned horizontally in relation to a
second column of the binary mirrors; a first lens operable to
collimate light from the first light source such that the first
column of mirrors is illuminated and further operable to collimate
light from the second light source such that the second column of
mirrors is illuminated; and a second lens operable to image light
from the first and second light sources horizontally.
34. The optical digital signal processing system of claim 33,
wherein the plurality of detectors comprises a first detector
positioned vertically in relation to a first row of the binary
mirrors and a second detector positioned vertically in relation to
a second row of the binary mirrors and the optical digital signal
processing system further comprises: a third lens operable to
collimate light from the first row of binary mirrors such that the
first detector is illuminated and to collimate light from the
second row of binary mirrors such that the second detector is
illuminated; and a fourth lens operable to image light from the
first and second row of mirrors vertically.
35. The optical digital signal processing system of claim 27,
wherein the light source comprises a vertical cavity surface
emitting laser diode.
36. The optical digital signal processing system of claim 27,
wherein the digital signal processor comprises a first digital
signal processor and the optical digital signal processing system
further comprises: an analog-to-digital converter coupled to the
detector array; and a second digital signal processor coupled to
the analog-to-digital converter.
37. The optical digital signal processing system of claim 27,
wherein the digital micromirror device comprises an array of
768.times.1024 binary mirrors.
38. The optical digital signal processing system of claim 27,
wherein the subset of the binary mirrors are optically linked to a
substantially similar number of detectors.
39. The optical digital signal processing system of claim 27,
wherein the digital micromirror device further comprises a
perimeter of guard pixels.
40. The optical digital signal processing system of claim 27,
further comprising a beamsplitter operable to receive light from
the light source and direct the light to the digital micromirror
device and further operable to receive the illumination from the
digital micromirror device and direct the illumination to the
detector array.
41. A method for optical digital signal processing, comprising:
configuring a plurality of binary pixels to allow a subset of the
binary pixels to represent a range of values; receiving light from
a light source at the plurality of binary pixels; altering the
intensity of the light to represent one of the values based, at
least in part, on the configuration of the subset of the binary
pixels; and transmitting the altered light from the plurality of
binary pixels to a detector array.
Description
TECHNICAL FIELD OF THE INVENTION
[0001] This invention relates generally to optical signal
processing and, more specifically, to optical digital signal
processing.
BACKGROUND OF THE INVENTION
[0002] Current optical signal processing systems include input
modulation that is analog. This discrete analog input is used to
drive the input light source. Conventional systems use analog
spatial light modulators (SLMs) and therefore the output is also
analog. Further, the detector array is also analog,often with
limited precision, and normally requires its output to be digitized
via an analog-to-digital (A/D) converter. The precision of the
digitized output is then limited to the number of bits of the A/D
dynamic range (e.g. 12.about.16 bits). Accordingly, precision loss
may occur during A/D conversion. On the other hand, conventional
digital SLMs comprised of binary micromirrors require 65,536
(2{circumflex over ( )}16) mirrors to represent a sixteen bit
number.
SUMMARY OF THE INVENTION
[0003] In accordance with the present invention, the disadvantages
and problems associated with analog and binary digital SLMs have
been substantially reduced or eliminated.
[0004] One aspect of the invention is a method for optical digital
signal processing, that comprises configuring a plurality of binary
mirrors to allow a subset of the binary mirrors to represent a
range of values. The plurality of binary mirrors comprises adigital
micromirror device. Light from a light source is received at the
digital micromirror device. The intensity of the light is altered
to represent one of the values based, at least in part, on the
configuration of the subset of the binary mirrors. The altered
light is transmitted from the digital micromirror device to a
detector array.
BRIEF DESCRIPTION OF THE DRAWINGS
[0005] For a more complete understanding of the present invention
and its advantages, reference is now made to the following
descriptions, taken in conjunction with the accompanying drawings,
in which:
[0006] FIG. 1 is a diagram of an optical digital signal processing
system in accordance with one embodiment of the present
invention;
[0007] FIGS. 2A-C illustrate alternative embodiments of the optical
digital signal processing system of FIG. 1;
[0008] FIGS. 3A-B illustrate operations of the lenses in accordance
with the parallel optical digital signal processing system of FIG.
2C;
[0009] FIGS. 4A-D illustrate various embodiments of a Radix-4
subset in accordance with the optical digital signal processing
system of FIG. 1;
[0010] FIGS. 5A-C illustrate various embodiments of a Radix-16
subset in accordance with the optical digital signal processing
system of FIG. 1;
[0011] FIGS. 6A-D illustrate various arrangements for representing
analog values to improve detectability by the detector array in
accordance with the Radix subsets of FIG. 5A and FIG. 4A; and
[0012] FIG. 7 illustrates a portion of the digital micromirror
device in accordance with the optical digital signal processing
system of FIG. 1.
DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS OF THE INVENTION
[0013] FIG. 1 illustrates one embodiment of an optical digital
signal processing system 10. Other embodiments of optical digital
signal processing system 10 may be formed without departing from
the scope of this disclosure. In general, the present invention
offers optical digital signal processing with integration,
resolution, accuracy, and programmability not afforded by analog
optical systems.
[0014] Optical digital signal processing system 10 includes light
source 20, digital micromirror device (DMD) 30, detector array 40,
first digital signal processor 50, second digital signal processor
60, analog-to-digital (A/D) converter 70, and digital-to-analog
(D/A) converter 80. Light source 20, digital micromirror device 30,
and detector array 40 are optically connected. Each component may
be optically coupled through lenses, collimators, or other suitable
optical devices, as illustrated in FIGS. 2A-C. Light source 20 is a
device that emits light energy 25 in the direction of DMD 30 to
initiate optical signal processing. Light source 20 may include one
or more light-emitting diodes (LEDs), injection laser diodes
(ILDs), Vertical Cavity Surface Emitting Lasers (VCSEL), an array
of light sources, or any other appropriate device operable to emit
light energy 25. Light source 20 is controlled by digital signal
processor 50.
[0015] Digital signal processor 50 is coupled to light source 20
and DMD 30. In certain embodiments, digital signal processor 50 is
coupled to light source 20 through D/A converter 80. Digital signal
processor 50 may be a reduced-functionality digital signal
processor, such that there are no multipliers present in digital
signal processor 50. Therefore, reduced-functionality digital
signal processor 50 may have much less power dissipation than
traditional digital signal processors. According to particular
embodiments, digital signal processor 50 generates and digitizes an
input sequence that drives light source 20. The generated analog
input signal may be digitized via pulse-code-modulation (PCM) or
pulse-width-modulation (PWM) to produce an input signal with
digital precision. If present, D/A converter 80 converts the
digital input signal to an analog control signal, which is
communicated to and controls light source 20. It should be
understood that if the input sequence was digitized using PWM, then
D/A converter 80 is not necessary. Digital signal processor 50 is
further operable to configure and program DMD 30 to allow DMD 30 to
process analog optical signals with digital precision.
[0016] Digital micromirror device (DMD) 30 is a digital form of a
spatial light modulator that acts as a matrix mask configured by
digital signal processor 50. In general, DMD 30 processes energy 25
received from light source 20 into an image 35, based on the
configuration, and transmits the resulting image 35 to detector
array 40. In one embodiment, DMD 30 is an electromechanical device
including a pixel array, such as a 768.times.1024 array, of digital
tilting mirrors or baseline binary pixels or mirrors. Each binary
mirror may tilt by a plus or minus angle (e.g. ten or twelve
degrees) for the active "on" or "off" positions. To permit the
mirrors to tilt, each is attached to one or more actuators such as,
for example, hinges mounted on support posts over underlying
control circuitry. The control circuitry provides electrostatic
forces that causes each mirror to selectively tilt. Incident light
on the mirror array is reflected by the "on" mirrors in one
direction and by the "off" mirrors in the other direction. The
configurable pattern of "on" versus "off" mirrors forms image 35.
Accordingly, image 35 is energy that is reflected by the "on"
mirrors in DMD 30 and generally projected to detector array 40. It
should be understood that system 10 may use "off" mirrors instead
of "on" mirrors to form image 35 to achieve the desired intensity
of image 35.
[0017] Detector array 40 may be a photo-detector array (PDA), a
charge-coupled device (CCD) detector array, CMOS array imager, or
any other suitable image detection device operable to obtain a
spectral response from image 35. In general, detector array 40
includes a plurality of detectors that receive image 35 and
converts the constituent optical signals into electrical signals
that may be further processed by components of system 10.
Accordingly, detector array 40 converts the energy from image 35
into a series of successive electrical pulses, which are output to
digital signal processor 60.
[0018] Digital signal processor 60 is coupled with detector array
40 through A/D converter 70. A/D converter 70 converts the analog
input signal from detector array 40 to a multi-bit digital output
signal, which is supplied to digital signal processor 60. Further,
digital signal processor 60 may be a reduced-functionality digital
signal processor, such that there are no multipliers present in
digital signal processor 60. Therefore, reduced-functionality
digital signal processor 60 may have much less power dissipation
than traditional digital signal processors.
[0019] It should be understood that while FIG. 1 illustrates light
source 20, DMD 30, detector array 40, first digital signal
processor 50, second digital signal processor 60, A/D converter 70,
and D/A converter 80 as separate components of system 10, one or
more of these components may be integrated into a single module or
chip. For example, DMD 30, digital signal processor 50 and D/A
converter 80 may be formed on a single chip. Alternatively, or in
addition, detector array 40, second digital signal processor 60,
and A/D converter 70 may be formed on a second chip.
[0020] In one aspect of operation, digital signal processor 50
configures DMD 30 such that optical signal processing may be
performed with digital precision. Accordingly, digital signal
processor 50 configures DMD 30 to allow one or more subsets of
binary mirrors to each represent a range of values. Digital signal
processor 50 may dynamically configure DMD 30 in order to
electronically process digital signals based on various optical
properties. For example, digital signal processor 50 may configure
a first subset of four binary mirrors to represent a range of
values between "0" and "3". In another example, digital signal
processor 50 may configure a first subset of sixteen binary mirrors
to represent a range of values between "0" and "15". System 10 may
also allow for the represented range of values to include negative
values. The negative values may be represented through any
appropriate technique such as, for example, twos complement, least
significant bit-inversion, redundant signed digits, or any higher
radix representation. For example, digital signal processor 50 may
configure the subset of binary mirrors such that a photonegative
pattern may be reflected. According to particular embodiments, this
photonegative pattern may be the reverse of another pattern. For
example, if a first pattern includes the first mirror in the "on"
position and the remaining mirrors in the "off" position, then the
photonegative pattern of the first pattern may include the first
mirror in the "off" position and the remaining mirrors in the "on"
position. It should be understood that system 10 may assign each
"on" mirror a value of 1, with each "off" mirror assigned a value
of 0 or use any other appropriate valuation scheme or
technique.
[0021] Based on various configurations, each subset may each be
assigned a mathematical weight to allow each subset to represent a
different portion of a single datum, which may be based on the
number of mirrors per subset. For example, a first subset of
sixteen mirrors may represent a first nibble of data (16{circumflex
over ( )}0) and a second subset of sixteen mirrors may represent a
second nibble of data (16{circumflex over ( )}1). The configuration
process may further include aligning light source 20 with DMD 30.
Generally, aligning light source 20 with DMD 30 may include
"chasing" emitted light energy 25 on DMD 30. For example, digital
signal processor 50 instructs light source 20 to emit a short beam
of light energy 25 to a selected first subset of binary mirrors
that represents the range of values. In response to a second subset
being illuminated by energy 25, digital signal processor selects
the second subset to represent the range of values and replaces the
first subset with the selected second set.
[0022] Once digital signal processor 50 configures DMD 30, system
10 may use the various components for optical digital signal
processing. To initiate an optical signal, digital signal processor
50 generates an input sequence that drives light source 20. Digital
signal processor 50 may digitize the generated input sequence and
communicate the digital input sequence to light source 20. In one
embodiment, the communicated digital input sequence is processed by
D/A converter 80, which converts the digitized sequence into an
analog control signal that drives light source 20.
[0023] Based upon the input sequence, light source 20 emits energy
25 towards at least a portion of DMD 30. Generally, DMD 30 receives
energy 25 and digitally forms a reflected image 35 for subsequent
numerical processing. Accordingly, DMD 30 receives commands from
digital signal processor 50 and appropriately arranges none, some,
or all of the binary mirrors in each subset to be in the "on"
position. Energy 25 is reflected by the "on" mirrors towards
detector array 40 and forms image 35. For example, DMD 30 may
activate none of a first set of binary mirrors, such that the
subset represents the analog value "0". DMD may then activate three
binary mirrors in a second subset to be in the "on" position, which
would represent the analog value "3". These reflections from the
"on" binary mirrors, digitally representing the analog values,
produce image 35 with digital precision.
[0024] Image 35 is received by one or more detectors of detector
array 40 for conversion and subsequent processing. Detector array
converts the optical signals from image 35 into analog electrical
signals and communicates the electrical signals to digital signal
processor 60 through A/D converter 70. A/D converter 70 converts
each analog signal into a multi-bit digital signal and transmits
the resulting digital signal to digital signal processor 60. As
image 35 was formed with digital techniques, there should be
substantially less precision loss during conversion by A/D
converter 70. Digital signal processor 60 receives and processes
the converted signals using any appropriate digital signal
processing technique.
[0025] FIG. 2A illustrates an example embodiment of a serial
optical digital signal processing system 201. In general, system
201 processes beams of energy 25 to produce image 35 in a serial,
or one-by-one, fashion.
[0026] System 201 includes collimator 100 and lens 110. Collimator
100 directs energy 25 to the appropriate portion of DMD 30, which
may include one or more subsets of binary mirrors. Collimator 100
may be any optical collimator with an optical scheme that can
produce a beam of parallel rays of energy. Energy 25 is directed to
DMD 30 and is reflected from mirrors 34 that are in the "off"
position to form a second image 37. Image 37 is directed away from
detector array 40 such that image 37 does not distort image 35.
Energy 25 that reflects from mirrors 32, which are activated to be
in the "on" position, forms image 35 and is directed towards
detector array 40 through lens 110.
[0027] Lens 110 is any lens and/or mirror that directs the line of
sight of image 35 to detector array 40 and, further, may be
multi-spectral refractive or reflective. Lens 110 allows
oversampling of the image 35 from DMD 30 to detector array 40
through optically linking each subset of binary mirrors to a single
detector in detector array 40. For example, lens 110 may focus the
portion of image 35 from one subset of sixteen mirrors to one
detector. In another example, if image 35 of the sixteen mirrors is
now focused onto an area of sixteen detectors then an oversampling
of image 35 of sixteen times is achieved for more precision. In
general, the use of oversampling may allow for a significantly
higher detection precision or the use of a much smaller detector
array 40 with fewer detectors.
[0028] In one aspect of operation, light source 20 emits energy 25
towards DMD 30 through collimator 100. Collimator 100 bends light
energy 25 to focus on the desired portion of DMD 30. "On" mirrors
32 reflect the directed energy 25 to form image 35. Image 35 is
focused towards the desired portion of detector array 40 using lens
110.
[0029] FIG. 2B illustrates another embodiment of a serial optical
digital signal processing system 202. Similar to system 201, system
202 performs serial optical digital signal processing. In addition
to the components of system 201, system 202 includes beamsplitter
120 that allows for less overall space being required for system
10. Beamsplitter 120 is capable of splitting energy 25 from image
35 and may include a beamsplitter, a reflector plate that guides
beams of light, or any mirror or prism or a combination of the two
that is used to divide energy into two or more parts. Beamsplitter
120 simultaneously reflects, or directs, energy 25 to DMD 30 and
allows passage of image 35 to detector array 40.
[0030] In one aspect of operation, light source 20 emits energy 25
towards DMD 30 through collimator 100. Collimator 100 bends light
energy 25 to focus on the desired portion of beamsplitter 120.
Beamsplitter 120 reflects the substantially all of energy 25
towards DMD 30. "On" mirrors 32 reflect the directed energy 25 to
form image 35. Image 35 travels through beamsplitter 120 and is
focused towards the desired portion of detector array 40 using lens
110.
[0031] FIG. 2C illustrates an example embodiment of parallel
optical digital signal processing system 203. In general, system
203 processes beams of energy 25 to produce image 35 in a parallel
fashion allowing the optical processing of vector inner products.
In this embodiment, light source 20 includes a plurality of input
diodes horizontally positioned to DMD 30 such that each input diode
can transmit to an entire column of subsets of mirrors. According
to one embodiment, the number of input diodes should substantially
equal the number of columns of subsets. Also, detector array 40 is
vertically positioned to DMD 30 such that the portion of image 35
reflected from each row of subsets can be summed by a single
detector.
[0032] System 203 includes lens combination 120 and lens
combination 140. Lens combination 120 spreads energy 25 from each
input diode vertically to the respective column of subsets of
binary mirrors associated with DMD 30. According to a particular
embodiment, lens combination 120 includes a spherical lens followed
by a cylindrical lens that has no power in the vertical direction.
Therefore, lens combination 120 collimates light diverging
vertically while directing light horizontally to the entire column
of subsets, as illustrated in FIG. 3A.
[0033] Lens combination 140 reduces the portion of image 35 from an
entire row of subsets to a single detector in detector array 40.
According to a particular embodiment, lens combination 140 includes
a spherical lens followed by a cylindrical lens that has no power
in the horizontal direction. Consequently, light from each row of
subsets is focused horizontally but directed vertically to the
respective detector as illustrated in FIG. 3B. Therefore, the light
from input source 20 is spread vertically such that each detector
detects a vector inner product of the light.
[0034] The preceding illustrations and accompanying descriptions
provide exemplary diagrams for implementing various optical digital
signal processing schemes. However, these figures are merely
illustrative, and system 10 contemplates using any suitable
combination and arrangement of elements for implementing various
optical digital signal processing schemes. Thus, these systems may
include any suitable combination and arrangement of elements for
processing energy 25 to produce an image 35 for digital signal
processing. Moreover, the operations of the various illustrated
systems may be combined and/or separated as appropriate. For
example, optical digital signal processing 10 may include
components from example serial systems 201 and 202 and parallel
system 203 as appropriate.
[0035] FIGS. 4A-D illustrate various embodiments of a Radix-4
subset in accordance with optical digital signal processing system
10. In general, the Radix-4 subset is a four-mirror subset of the
plurality of binary mirrors included in DMD 30. The Radix-4 subset
allows optical digital signal processing system 10 to configure
four binary mirrors to represent a range of four analog values such
as, for example, "0"-"3". Illustrated in FIG. 4A is a square
Radix-4, or dibit, subset 302. Radix-4 subset 302 includes a first,
second, third, and fourth binary mirror. In this example
embodiment, the first and second mirrors are diagonally positioned
in relation to one another. Further, the third and fourth binary
mirrors are diagonally positioned from one another and further
positioned in relation to the first and second binary mirrors to
form a 2.times.2 square pattern. Illustrated in FIG. 4B, are four
similar binary mirrors that are horizontally positioned in relation
to one another creating a 1.times.4 binary mirror subset.
Illustrated in FIG. 4C, are four similar binary mirrors vertically
positioned in relation to one another to form a 4.times.1
rectangular binary mirror subset.
[0036] FIG. 4D illustrates one aspect of operation of the Radix-4
square pattern 302, as illustrated in FIG. 3A. As illustrated in
FIG. 4D, the Radix-4 subset may represent five values based upon
whether the individual binary mirrors are in the "on" or "off"
positions. For example, if all four binary mirrors are in the "off"
position, optical digital signal processing system 10 may interpret
this subset of binary mirrors as representing an analog value "0".
Activating the first binary mirror to be in the "on" position,
changes the analog value represented by the subset to "1". Further
activating the second binary mirror changes the analog value of the
subset to "2". Additionally, activating the third binary mirror
changes the analog dibit value to "3". According to particular
embodiments, activating the fourth binary mirror, resulting in all
four binary mirrors being in the "on" position, illustrates a test
digit that may be used to verify proper functioning of optical
digital signal processing system 10 or a numerical overflow signal
for processing vectors.
[0037] FIGS. 5A-C illustrate various embodiments of a Radix-16
subset in accordance with optical digital signal processing system
10. In general, the Radix-16 subset is a sixteen-mirror subset of
the plurality of binary mirrors included in DMD 30. The Radix-16
subset allows optical digital signal processing system 10 to
configure sixteen binary mirrors to represent a range of sixteen
analog values such as, for example, "0" - "15". Illustrated in FIG.
5A, is a 4.times.4 Radix-16, or HEX, square subset 402. Illustrated
in FIG. 5B, is an 8.times.2 Radix-16 subset 404. Radix-16 subset
404 includes a first column of eight binary mirrors, each
positioned vertically in relation to the other seven. Radix-16
subset 404 further includes a second column of eight binary
mirrors, each also positioned vertically in relation to one another
in the second column. Illustrated in FIG. 5C, is a 2.times.8
Radix-16 subset 406. Radix-16 subset 406 includes a first row eight
binary mirrors, each positioned horizontally in relation to the
other seven binary mirrors. Radix-16 subset 406 further includes a
second row of eight binary mirrors, each binary mirror also
positioned horizontally in relation to the other seven in the
second row.
[0038] FIGS. 6A-D illustrate various arrangements for representing
analog values to improve detectability by detector array 40 in
accordance with the Radix-16 subset 402 of FIG. 5C and the Radix-4
subset 302 of FIGS. 4A. FIG. 6A illustrates square Radix-16 subset
402 that is configured to represent sixteen analog values such as,
for example, "0" - "15". In the illustrated embodiment, the shaded
squares illustrate a binary mirror in the "off" position, while the
light squares represent a binary mirror in the "on" position. As
described above, digital signal processor 50 arranges the subset of
binary mirrors to represent one of the range of values. As
illustrated, if digital signal processor 50 determines that this
subset should represent the analog value "0", then digital signal
processor 50 communicates the appropriate signal to DMD 30. In
response to the signal, DMD 30 insures that all sixteen mirrors are
arranged in the "off" position. It should be understood that to
achieve the desired arrangement, DMD 30 may switch the appropriate
binary mirrors from the "on" to the "off" position and leave the
remaining binary mirrors in the "off" position.
[0039] In response to a signal requesting the analog value "1", DMD
30 selects a first mirror from the subset and activates the mirror
to be in the "on" position. According to particular embodiments,
the first mirror is located in a central position in the subset to
improve detectability by detector array 40. In response to a signal
requesting the analog value "2", DMD 30 selects a second mirror
from the subset. According to particular embodiments, DMD 30
selects the second mirror that is diagonally positioned to the
first mirror. This may improve detectability by detector array 40
by decreasing the chances of overlap of the portion of image 35
transmitted by the first and second mirror. It may further decrease
the probability of overlap between the illustrated subset and a
second subset (illustrated in more detail in FIG. 7). In response
to signals for increased analog values, DMD 30 continues to select
and activate additional binary mirrors, to be in the "on" position,
in locations that improve detectability by detector array 40 over
other arrangements. Similar to FIG. 4D, when all sixteen mirrors of
the subset are activated to be in the "on" position, the subset may
represent a test digit or numerical overflow.
[0040] FIG. 6B illustrates three example arrangements of Radix-4
subsets 302 for numeric processing. Each example includes four
subsets of four binary mirrors, illustrated as square pattern
Radix-4 subset 302. It should be understood that this is for
illustration purposes only and that any number of mirrors, in any
configuration, may be used for signal processing.
[0041] Returning to the example, the first Radix-4 subset 302
includes three of four mirrors activated to be in the "on" position
to represent the analog value "3". The second Radix-4 subset 302
has two of the four binary mirrors activated to be in the "on"
position in order to represent the analog value "2". The third
Radix-4 subset 302 has one mirror activated to be in the "on"
position in order to represent the analog value "1". The fourth
Radix-4 subset 302 has no mirrors activated to be in the "on"
position and represents the analog value "0". Each subset has been
given a weight based on the power of four, which is based on the
number of mirrors per subset. For illustration purposes only, the
first Radix-4 subset 302 is assigned a weight based on four to the
power of zero. The second Radix-4 subset 302 is assigned the weight
four to the power of one. The third Radix-4 subset 302 is assigned
the weight four to the power of two. The fourth Radix-4 subset 302
is assigned the weight four to the power of three. This results in
a numerical value of "27" and is represented as "0123" dibit or
"1B" HEX. As illustrated, the four Radix-4 subsets may be arranged
in a square (2.times.2) pattern 302, a rectangular (1.times.4)
pattern 304, or a rectangular (4.times.1) pattern 306.
[0042] FIG. 6C illustrates six example arrangements of Radix-16
subsets 402 for numeric processing. Each example includes three
subsets of sixteen binary mirrors, illustrated as square pattern
Radix-16 subset 402. It should be understood that this is for
illustration purposes only and that any number of mirrors, in any
configuration or pattern, may be used for signal processing.
[0043] Returning to the example, the first Radix-16 subset 402
includes three of sixteen mirrors activated to be in the "on"
position to represent the analog value "3". The second Radix-16
subset 402 has two of the sixteen binary mirrors activated to be in
the "on" position in order to represent the analog value "2". The
third Radix-16 subset 402 has one mirror activated to be in the
"on" position in order to represent the analog value "1". Each
subset has been given a weight based on the power of sixteen, which
is based on the number of mirrors per subset. For illustration
purposes only, the first Radix-16 subset 402 is assigned a weight
based on sixteen to the power of zero. The second Radix-16 subset
402 is assigned the weight sixteen to the power of one. The third
Radix-16 subset 402 is assigned the weight sixteen to the power of
two. This results in a numerical value of "0123" HEX. As
illustrated, the three Radix-16 subsets may be arranged in a
rectangular (1.times.3) pattern, a rectangular (3.times.1) pattern,
or various other patterns. FIG. 6D illustrates three example
arrangements of Radix-16 subsets 402 for numeric processing. Each
example includes four subsets of sixteen binary mirrors,
illustrated as square pattern Radix-16 subset 402. It should be
understood that this is for illustration purposes only and that any
number of mirrors, in any configuration, may be used for signal
processing.
[0044] Returning to the example, the first Radix-16 subset 402
includes three of sixteen mirrors activated to be in the "on"
position to represent the analog value "3". The second Radix-16
subset 402 has two of the sixteen binary mirrors activated to be in
the "on" position in order to represent the analog value "2". The
third Radix-16 subset 402 has one mirror activated to be in the
"on" position in order to represent the analog value "1". The
fourth Radix-16 subset 402 has no mirrors activated to be in the
"on" position and represents the analog value "0". Each subset has
been given a weight based on the power of sixteen, which is based
on the number of mirrors per subset. For illustration purposes
only, the first Radix-16 subset 402 is assigned a weight based on
sixteen to the power of zero. The second Radix-16 subset 402 is
assigned the weight sixteen to the power of one. The third Radix-16
subset 402 is assigned the weight sixteen to the power of two. The
fourth Radix-16 subset 402 is assigned the weight sixteen to the
power of three. This results in a numerical value of "0123" HEX. As
illustrated, the four Radix-16 subsets may be arranged in a square
(2.times.2) pattern, a rectangular (1.times.4) pattern, a
rectangular (4.times.1) pattern, or any other appropriate pattern
that may improve detectability or processing.
[0045] FIG. 7 illustrates an example portion of DMD 30 in
accordance with optical digital signal processing system 10. This
exemplary portion of DMD 30 includes four square Radix-16 subsets
402 and is substantially delimited by guard pixels 36. Each guard
pixel 36 is a binary mirror, which is not activated, that is not
configured to be included in one of the subsets. This allows system
10 to use dead space to ensure that the portion of image 35
reflected from each subset 402 does not interfere with or corrupt
other portions of image 35. The illustrated portion of DMD 30 is
surrounded by a perimeter of guard pixels 36. Further, each subset
402 is separated from each other by a row or column of guard pixels
36, as appropriate.
[0046] Although the present invention has been described in detail,
it should be understood that various changes, substitutions and
alterations can be made hereto without departing from the sphere
and scope of the invention as defined by the appended claims.
[0047] To aid the Patent Office, and any readers of any patent
issued on this application in interpreting the claims appended
hereto, applicants wish to note that they do not intend any of the
appended claims to invoke .paragraph. 6 of 35 U.S.C. .sctn. 112 as
it exists on the date of filing hereof unless "means for" or "step
for" are used in the particular claim.
* * * * *