FCD4A14CCB [ATMEL]
Image Sensor, 1 Func, CMOS, COB-21;
| 型号： | FCD4A14CCB |
| 厂家： | 
ATMEL |
| 描述： | Image Sensor, 1 Func, CMOS, COB-21 |
| 文件： | 总19页 (文件大小：356K) |
| 中文： | 中文翻译 | 下载： | 下载PDF数据表文档文件 |
Features
• Sensitive Layer Over a 0.8 mm CMOS Array
• Image Zone: 0.4 x 14 mm = 0.02" x 0.55"
• Image Array: 8 x 280 = 2240 pixels
• Pixel Pitch: 50 mm x 50 mm = 500 dpi
• Pixel Clock: up to 2 MHz Enabling up to 1780 Frames per Second
• Die Size: 1.7 x 17.3 mm
• Operating Voltage Range: Nominal 3V to 5.5V
• Power Consumption: 20 mW @ 3.3V, 1 MHz, 25°C
• Operating Temperature Range: 0°C to +70°C: C suffix
(-40°C to +85°C to be characterized)
• Naturally Protected Against ESD: > 16 kV Air Discharge
• Resistant to Abrasion: >1 Million Finger Sweeps
• 20-lead Ceramic DIP or Chip-On-Board (COB) Package, with Specific Protective Layer
Thermal
Fingerprint
Sensor with
0.4 mm x 14 mm
(0.02" x 0.55")
Sensing Area
and
Applications
• Terminal Access (PCs, access to networks, etc.)
• Electronic payment associated with payment card (Auto-mated Teller Machine,
Portable Point Of Sale, etc.)
• Building access
• Electronic keys (cars, home, etc.)
• Cellular phones (usable only by registered users)
• Portable fingerprint imaging for law enforcement
• TV access
• Weapons (usable only by registered users)
Digital Output
(on-chip ADC)
Step for easy
integration
Chip-on-Board Package
(COB)
Sensing area
FDC4A14
Wire protection
(not drawn)
FingerChip™
20-pin, 0.3" Dual-Inline
Ceramic Package
Actual Size
(DIP20)
Description
FCD4A14 is part of the FingerChip™ Atmel monolithic fingerprint sensor family for
which no optics, no prism and no light source are required.
FCD4A14 is a single chip, high performance, low cost sensor based on temperature
physical effects for fingerprint sensing.
FCD4A14 has a linear shape, allowing for the capture of a fingerprint image by
sweeping the finger across the sensing area. After capturing several images, Atmel
proprietary software can reconstruct a full 8-bit fingerprint image, if needed.
FCD4A14 has a small surface combined with CMOS technology, and a ceramic dual-
in-line or Chip-On-Board package assembly. These facts contribute to a low-cost
device.
Rev. 1962A–01/00
FCD4A14 delivers a programmable number of images per
second, while an integrated Analog to Digital Converter
delivers a digital signal adapted to interfaces such as an
EPP parallel port, USB microcontroller or directly to micro-
processors. Thus, no frame grabber or glue interface is
necessary to send the frames. These facts make FCD4A14
an easy device to include in any system for identification or
verification applications.
Absolute Maximum Ratings (1)
Parameter
Symbol
VCC
Comments
Value
Unit
V
Positive supply voltage
Temperature stabilization power
Front plane
GND to 6.5
GND to 6.5
GND to VCC
GND to VCC
TPP
V
FPL
V
Digital input voltage
Digital output current
Die temperature
RSTPCLK
ID
V
mA
°C
°C
Tj
-55 to +85
-55 to +85
Storage temperature
Tstg
Do not solder
DIP: socket mandatory
Lead temperature (soldering 10 s)
Tleads
Forbidden
°C
Note:
1. Absolute maximum ratings are limiting values, to be applied individually, while other parameters are within specified operat-
ing conditions. Long exposure to maximum ratings may affect device reliability.
Recommended Conditions Of Use
Parameter
Symbol
Comments
Min
Typ
5V
Max
Unit
V
Positive supply voltage
Front plane
VCC
3V
5.5V
FPL
Must be grounded.
GND
V
Digital input voltage
Digital output voltage
Digital load
CMOS levels
CMOS levels
50
V
V
CL
pF
CA
RA
20
10
pF
Analog load
kΩ
Operating temperature range
Duty cycle
Tamb
DC
Civil: “C” grade
0 to +70
50
°C
Clock PCLK
20
80
%
FCD4A14
2
FCD4A14
Resistance
Min value
Standard method
ESD
On pins. HBM (Human Body Model) CMOS I/O
2 kV
MIL-STD-883- method 3015.7
On die surface (Zapgun)
Air discharge
Contact
16 kV
9 kV
NF EN 6100-4-2
CEI 1000-4-2
MECHANICAL ABRASION
# cycles without lubricant multiply by a factor of 20
for correlation with a real finger
300 000
4 hours
MIL E 12397B
CHEMICAL RESISTANCE
Cleaning agent, acid, grease, alcohol, diluted
acetone
Internal method
Specifications
Test
Parameter
Symbol
Tamb
level
Min
Typ
50
Max
Unit
micron
pixel
Resolution
IV
IV
I
Size
8x280
Yield: number of bad pixels
Equivalent resistance on TPP pin
15
bad pixels
ohm
I
30
Explanation Of Test Levels
I
100% production tested at +25°C
II
100% production tested at +25°C, and sample tested at specified temperatures (AC testing done on sample)
Sample tested only
III
IV
V
VI
D
Parameter is guaranteed by design and/or characterization testing
Parameter is a typical value only
100% production tested at temperature extremes.
100% probe tested on wafer at Tamb = +25°C
3
5 Volt
Power supply = +5V; Tamb = 25°C; FPCLK = 1 MHz; Duty cycle = 50%;
Cload 120 pF on digital outputs, analog outputs disconnected otherwise specified.
Parameter
Symbol
Tamb
Test level
Min
Typ
Max
Unit
Power Requirements
Positive supply voltage
VCC
4.5
5
5.5
V
Digital positive supply current on VCC pin
Cload = 0
I
13
6
mA
mA
ICC
IV
Power dissipation on VCC
Cload = 0
I
65
30
mW
mW
PCC
IV
Power dissipation on VCC in NAP mode
Analog Output
PCCNAP
I
I
0.1
2.8
mW
Voltage range
VAVx
0
V
Digital Inputs
Logic compatibility
CMOS
Logic “0” voltage
VIL
VIH
IIL
I
I
I
I
0
1.2
VCC
100
100
V
V
Logic “1” voltage
3.6
Logic “0” current
µA
µA
Logic “1”current
IIH
Digital Outputs Cload 120 pF
Logic compatibility
CMOS
Logic “0” voltage(1)
VOL
VOH
tr
I
1.5
V
V
Logic “1” voltage(1)
I
3.5
Output rise time (10% - 90% final value)
Output rise time (10% - 90% final value)
IV
IV
10
5
ns
ns
tf
Note:
1. With IOL = 1 mA and IOH = -1 mA
FCD4A14
4
FCD4A14
.
3.3 Volt
Power supply = +3.3V; Tamb = 25°C; FPCLK = 1 MHz; Duty cycle = 50%;
Cload 120 pF on digital outputs, analog outputs disconnected otherwise specified
Parameter
Symbol
Tamb
Test level
Min
Typ
Max
Unit
Power Requirements
Positive supply voltage
VCC
ICC
3.0
3.3
3.6
V
Digital positive supply current on VCC pin
Cload= 0
I
10
6
mA
mA
IV
Power dissipation on VCC
Cload = 0
I
33
20
mW
mW
PCC
IV
Power dissipation on VCC in NAP mode
Analog Output
PCCNAP
0.1
2.8
mW
Voltage range
VAVx
I
0
V
Digital Inputs
Logic compatibility
Logic “0” voltage
CMOS
VIL
VIH
IIL
I
I
I
I
0
0.8
VCC
100
100
V
V
Logic “1” voltage
2.3
Logic “0” current
µA
µA
Logic “1”current
IIH
Digital Outputs
Cload 120 pF
Logic compatibility
Logic “0” voltage(1)
Logic “1” voltage(1)
CMOS
VOL
VOH
tr
I
0.6
V
V
I
2.4
Output rise time (10% - 90% final value)
Output rise time (10% - 90% final value)
IV
IV
10
5
ns
ns
tf
Note:
1. With IOL = 1 mA and IOH = -1 mA
5
.
Switching Performances
Tamb = 25°C; FPCLK = 1 MHz; Duty cycle = 50%; Cload 120 pF on digital and analog outputs otherwise specified
Parameter
Symbol
FPCLK
THCLK
TLCLK
TSetup
THold
Tamb
Test level
Min
Typ
Max
2
Unit
MHz
ns
Clock frequency
I
I
I
I
I
0.1
1
Minimum clock pulse width (high)
Minimum clock pulse width (low)
Min clock setup time (high) / reset falling edge
Min clock hold time (high) / reset falling edge
250
250
0
ns
ns
20
ns
5.0 Volt
All power supplies = +5 V
Parameter
Symbol
TPLHACKN
TPHLACKN
TPDATA
Tamb
Test level
Min
Typ
20
Max
Unit
ns
Output delay from PCLK to ACKN rising edge
Output delay from PCLK to ACKN falling edge
Output delay from PCLK to Data output Dxi
Output delay from PCLK to Analog output Avx
Output delay from OE to data high-Z
Output delay from OE to data output
I
I
I
I
I
I
17
ns
68
ns
TPAVIDEO
TDATAZ
266
25
ns
ns
TZDATA
29
ns
3.3 Volt
All power supplies = +3.3 V
Parameter
Symbol
TPLHACKN
TPHLACKN
TPDATA
Tamb
Test level
Min
Typ
31
Max
Unit
ns
Output delay from PCLK to ACKN rising edge
Output delay from PCLK to ACKN falling edge
Output delay from PCLK to Data output Dxi
Output delay from PCLK to Analog output AVx
Output delay from OE to data high-Z
Output delay from OE to data output
I
I
26
ns
I
82
ns
TPAVIDEO
TDATAZ
I
266
34
ns
IV
I
ns
TZDATA
47
ns
FCD4A14
6
FCD4A14
Figure 1. Reset
THRST
Reset RST
THold
Clock PCLK
TSetup
Figure 2. Read One Byte / Two Pixels
FPCLK
THCLK
TLCLK
Clock PCLK
TPLHACKN
Acknowledge
ACKN
TPHLACKN
TPDATA
Data output
Data #N-1
Data #N
Do0-3, De0-3
Data #N+1
Video analog output
AVO, AVE
Data #N
TPAVIDEO
Figure 3. Output Enable
Output Enable
OE
TZDATA
TDATAZ
Hi-Z
Data output
Hi-Z
Data output
Do0-3, De0-3
7
Pin Connection For DIP Ceramic Package
GND
AVE
TPP
VCC
RST
OE
De0
De1
De2
1
2
3
4
5
6
7
8
9
20 AVO
19 TPE
18 PCLK
17 ACKN
16 GND
15 Do0
14 Do1
13 Do2
12 Do3
11 FPL
Pin number
Name
Type
1
2
GND
AVE
TPP
VCC
RST
OE
GND
Analog output
Power
3
4
Power
5
Digital input
Digital input
Digital output
Digital output
Digital output
Digital output
GND
6
7
De0
De1
De2
De3
FPL
Do3
Do2
Do1
Do0
GND
ACKN
PCLK
TPE
AVO
De3 10
8
9
10
11
12
13
14
15
16
17
18
19
20
Digital output
Digital output
Digital output
Digital output
GND
Digital output
Digital input
Digital input
Analog output
Die Attach is connected to pin 1 and 16, and must be grounded. FPL pin must be grounded.
Pad Connection For Chip-on-board Package
GND
AVE
AVO
TPP
TPE
VCC
GND
RST
PCLK
OE
ACKN 11
De0 12
Do0 13
De1 14
Do1 15
De2 16
Do2 17
De3 18
Do3 19
FPL 20
GND 21
1
2
Pad number
Name
Type
1
2
GND
AVE
AVO
TPP
TPE
VCC
GND
RST
PCLK
OE
GND
3
4
Analog output
Analog output
Power
5
3
6
4
7
5
Digital input
Power
8
6
9
10
7
GND
8
Digital input
Digital input
Digital input
Digital output
Digital output
Digital output
Digital output
Digital output
Digital output
Digital output
Digital output
Digital output
GND
9
10
11
12
13
14
15
16
17
18
19
20
21
ACKN
De0
Do0
De1
Do1
De2
Do2
De3
Do3
FPL
GND
GND
Die Attach is connected to pin 1, 7 and 21, and must be grounded. FPL pin must be grounded.
FCD4A14
8
FCD4A14
FCD4A14 Block Diagram
clock
reset
PCLK
RST
ACKN
column selection
1 dummy column
line sel
amp
even
odd
4
4 bit
ADC
De0-3
Do0-3
1
8
8
8 lines of 280 columns of pixels
latches
2240
8
4 bit
ADC
4
chip
temperature
sensor
chip temperature
stabilization
output
enable
analog
output
TPP
TPE
AVE AVO
OE
Pin Description
Name Pad
DIP
COB
Pin Type
Function
Ground
1
2
3
4
GND
VCC
PCLK
RST
4
18
5
1, 16 1,7,21 Ground
FPL
(pad11)
4
18
5
6
9
8
Power
Power supply
Pixel clock
Reset
DE3
DE2
DE1
DE0
OE
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(10) DO3
(9) DO2
(8) DO1
(7) DO0
(6) ACKN
(5) PCLK
(4) GND
(3) TPE
(2) AVO
(pad1) TO
Digital Input
Digital Input
17
RST
VCC
TPP
AVE
Data Output Enable. Tri-state when
high
5
OE
16
6
10
Digital Input
6
7
Do0
Do1
Do2
Do3
De0
De1
De2
De3
7
8
15
14
13
12
7
13
15
17
19
12
14
16
18
11
20
4
Digital Output
Digital Output
Digital Output
Digital Output
Digital Output
Digital Output
Digital Output
Digital Output
Digital Output
Ground
Odd pixel bit 0 LSB
TE (pad21)
Odd pixel bit 1
8
9
Odd pixel bit 2
9
10
15
14
13
12
6
Odd pixel bit 3 MSB
Even pixel bit 0 LSB
Even pixel bit 1
10
11
12
13
8
9
Even pixel bit 2
10
17
11
3
Even pixel bit 3 MSB
Acknowledge signal / EPP protocol
Front plane. Must be grounded.
Temp. stabilization power
Temp. stabilization enable
Analog output odd pixels
Analog output even pixels
Test purpose only
14 ACKN
15
16
17
18
19
20
21
FPL
TPP
TPE
AVO
AVE
TO
11
19
3
Power
19
20
2
5
Digital Input
Analog Output
Analog Output
Analog Input
Analog Input
2
3
20
1
2
N/A
N/A
N/A
N/A
TE
21
Test purpose only
N/A: not available
9
Die Mechanical Information
Mask set reference
H97A
Passivation/coating
Revision
Specific
A
Die size.
1.7 x 17.34 mm
100 x 100 µm
675 25 µm
Ti (front side)
Poly (back side)
Pad size
Back side potential
Transistor count
Die attach
Ground
18827
Epoxy
AlSi/Ti
Die thickness
Metallization
Bond wire
Then, as frames are recovered, the reconstruction routine
computes an 8-bit value for each pixel. This value is calcu-
lated from the pixels of each frame coming from the device,
which appear to be at the same place, therefore reducing
noise and increasing resolution.
Functional Description
The circuit is divided into two main sections: sensor and
data conversion. One particular column among 280+1 is
selected in the sensor array (1), then each pixel of the
selected column sends its electrical information to amplifi-
ers (2) (one per line), then two lines at a time are selected
(odd and even) so that two particular pixels send their infor-
mation to the input of two 4 bit Analog to Digital Converters
(3), so 2 pixels can be read for each clock pulse (4).
Start Sequence
Although a reset is normally needed only once, after power
up, it is better to reset the FingerChip before each finger-
print acquisition.
Sensor
A reset is not necessary between each frame acquisition!
Start sequence must consist of:
Each pixel is a sensor in itself. The sensor detects a tem-
perature differential between the beginning of acquisition
and the reading of information: this is the integration time.
The integration time begins with a reset of the pixel to a
predefined initial state. Note that the integration time reset
has nothing to do with the reset of the digital section.
1. Set the RST pin to high
2. Set the RST pin to low
3. Send 4 clock pulses (due to pipe-line)
4. Send clock pulses to skip the first frame.
Note that the first frame never contains relevant information
because the integration time is not correct.
Then, at a rate depending on the sensitivity of the pyroelec-
tric layer, on the temperature variation between the reset
and the end of the integration time, and on the duration of
the integration time, electrical charges are generated at the
pixel level.
Reading the Frames
A frame consists of 280 true columns + 1 dummy column of
8 pixels. As two pixels are output at a time, a system must
send 281x4 = 1124 clock pulses to read one frame.
Analog-to-Digital Converter /
Reconstructing an 8-bit Fingerprint
Image
Reset must be low when reading the frames.
Read One Byte / Output Enable
Clock is taken into account on the falling edge and data are
output on the rising edge.
An Analog to Digital Converter (ADC) is used to convert the
analog signal coming from the pixel into digital data that
can be used by a processor.
The U.S. Federal Bureau of Investigation requires a 256-
level grey scale image; that is, an 8 bit per pixel resolution.
For each clock pulse, after the start sequence, a new byte
is output on the Do0-3, De0-3 pins. This byte contains 2
pixels: 4 bit on Do0-3 (odd pixels), 4 bit on De0-3 (even
pixels).
As the data rate for parallel port and USB is in the range of
1 Megabyte per second, and we need at least a rate of 500
frames per second to reconstruct the image with a fair
sweeping speed for the finger, two 4-bit ADCs have been
used to output 2 pixels at a time on one byte.
To output the data, the output enable (OE) pin must be low.
When OE is high, the Do0-3 & De0-3 pins are in high
impedance state. This enables an easy connection to a
microprocessor bus without additional circuitry-it will enable
FCD4A14
10
FCD4A14
data output by using a chip select signal. Note that the
FCD4A14 is always sending data: there is no data
exchange to perform using read/write mode.
The 4 bytes of the dummy column contain a fixed pattern
on the two first bytes, and temperature information on the
last two bytes (see later):
Even
0000
0000
nnnn
pppp
Odd
1111
1111
rrrr
Video Output
Dummy Byte 1 DB1:
Dummy Byte 2 DB2:
Dummy Byte 3 DB3:
Dummy Byte 4 DB4:
An analog signal is also available on pins AVE & AVO.
Note that video output is available one clock pulse before
the corresponding digital output (one clock pipe-line delay
for the analog to digital conversion).
tttt
The sequence 00001111 00001111 is a very rare
sequence in real fingerprint image, because it means that
we have the sequence ridge/valley/ridge/valley within 4 pix-
els, that is 200mm. Moreover, this sequence appears on
every frame (exactly every 1124 clock pulses), so it is an
easy pattern to recognize for synchronization purposes.
Pixel Order
After a reset, pixel number one is located on the upper left
corner, looking at the chip with bond pads to the right. For
each column of 8 pixels, pixels 1-3-5-7 are output on odd
data Do0-3 pins, pixels 2-4-6-8 are output on even data
De0-3 pins. Most significant bit is bit #3, least significant is
bit #0.
Integration Time and Clock Jitter
The FCD4A14 is not very sensitive to clock jitter (clock vari-
ation). The most important requirement is a regular
integration time that ensures the frame reading rate is also
as regular as possible, in order to get consistent fingerprint
Synchronization: The Dummy Column
A dummy column has been added to the sensor to act as a
specific pattern to detect the first pixel. So, 280 true col-
umns + 1 dummy column are read for each frame.
Figure 4.
1
2
3
4
column selection
line sel
even
odd
4
4
4 bit
ADC
De0-3
Do0-3
8
8 lines of 280 columns of pixels
1 dummy column
amp
latches
8
4 bit
ADC
chip
temperature
sensor
Figure 5. Start Sequence
Reset RST
4+1124 clock pulses to skip the first frame
Clock PCLK
1
2
3
4
1
1124
1
11
Figure 6. Read One Frame
Reset RST is low
Column 1
Column 2
6
Column 280
Dummy Column 281
1
2
3
4
5
1119 1120 1121 1122 1123 1124
Clock PCLK
Pixels 1 & 2 3 & 4 5 & 6 7 & 8 1 & 2 3 & 4
7 & 8
DB1
DB2
DB3 DB4
Figure 7. Regular Integration Time
REGULAR INTEGRATION TIME
Frame n
Frame n+1
Frame n+2
Frame n+3
Clock PCLK
1124 pulses
1124 pulses
1124 pulses
1124 pulses
Figure 8.
Pixel #1 (1,1)
Pixel #2233(280,1)
Bondpads
Pixel #8 (1,8)
Pixel #2240(280,8)
Temperature Management
Each pixel of the FingerChip is a temperature sensor that
detects temperature changes. This change is generated by
the temperature difference between the finger and the chip.
The best case happens when there is a large temperature
difference, for instance when the chip temperature is very
low or very high: this is a rather unusual case. The worst
case happens when the finger temperature is exactly the
same as the chip temperature.
• First, read the sensor temperature. Most of the time, it
will be out of the critical area (near 33°C), so no
stabilization is required.
• If power is not a critical resource, and if the sensor
temperature is typically close to the critical area, the user
may always stabilize the temperature above the usual
finger temperature (>37°C). Note that this stabilization
may take one minute or more, depending on the
surroundings, thermal resistance and the chip’s initial
temperature.
In order to get a contrasted image, we need at least one
degree difference between the sensor and the finger. Criti-
cal temperature is in the range of 33°C ( 5 as finger
temperature may vary). Chip temperature stabilization cir-
cuitry is implemented on the FCD4A14 to stabilize the
sensor image quality when needed.
• If power is a critical resource, the best solution is to have
a first trial with the finger without stabilization as most of
the time the temperature difference will be high enough.
If authentication fails and we detect a chip temperature
that is in the critical area, simply enable the stabilization
feature and try again. It may take a few seconds, as we
just need to stabilize a few degrees above the measured
temperature.
Several strategies may be used depending on external
constraints:
FCD4A14
12
FCD4A14
Two separate features are available in the FCD4A14:
4. Set Output Enable OE pin to high, so current can be
drained through the outputs.
1. An absolute temperature sensor. Information is digi-
tally provided in the dummy bytes DB3 and DB4.
Nap mode
2. Temperature stabilization circuitry, with two pins,
TPP and TPE.
Reset RST
The stabilization feedback is externally managed: an exter-
nal processor or algorithm will decide whether or not this
feature is enabled. In this way, the user has full control over
power consumption.
Nap
Clock PCLK
TPP is the pin that delivers power, and must be externally
connected to the power supply through a resistor to limit
the maximum current and avoid reaching extreme tempera-
tures. Value of the resistor depends on external conditions
such as voltage, environmental use, thermal isolation…
Application Notes
Finger Speed Versus Acquisition Speed
A finger speed is:
• very very slow below 1 cm/s
TPE controls the injected power: when the temperature is
below the desired temperature, TPE must be set to high,
and when the temperature is reached, TPE must be set to
low. This is a digital input: no power is required to drive this
pin, so any processor output or bus may drive it.
• slow at a few cm/s (you have to take care to go slowly)
• normal at 10 cm/s
• fairly fast at 20 cm/s
Please contact Atmel for more information and assistance
in your specific application.
• maximum in the range of 100 cm/s (difficult to sweep)
As a full fingerprint image is reconstructed from the slices,
it is important that slices recover. The strict minimum is one
line of recovery, but 2 lines recovery is a good value, so the
finger must not move more than 6 pixels between two
frames.
Temperature information is digitally provided on DB3 &
DB4. Data format, values and a program to manage to
manage the temperature stabilization circuitry will be avail-
able once characterization of the chip is complete.
Maximum finger speed vs acquisition speed is summarized
in the following figure, and will help the user define the sys-
tem requirements for acquisition:
Power Management
Nap Mode
Acquisition Speed
Maximum Finger Speed
Comments
Several strategies are possible to reduce power consump-
tion when not in use.
kbyte/s
frame/s
89
cm/s
2.7
The simplest and most efficient is to cut the power supply,
using external means.
100
250
Slow
222
6.7
Normal/Bidir Parallel
Port
A nap mode is also implemented in the FCD4A14. To acti-
vate this nap mode, user must:
700
623
18.7
26.7
40.0
53.4
Fair Speed/EPP
Fast Speed/USB
Very Fast/Max USB
Extremely Fast
1. Set the reset RST pin to high. Doing this, all analog
sections of the device are internally powered down.
1000
1500
2000
890
2. Set the clock PCLK pin to high (or low), thus stop-
ping the entire digital section.
1335
1779
3. Set the TPE pin to low or disconnect TPP to stop
the temperature stabilization feature.
13
Power must be supplied via the PS/2 port, for instance, as
power is not consistently available on the parallel port.
EPP Parallel port
From a software point of view, the system must have a high
priority during acquisition, because if the PC does some-
thing else (such as accessing the hard drive), some frames
may be skipped resulting in a “hole” in the fingerprint
image.
FCD4A14
De0
De1
De2
De3
Do0
Do1
Do2
Do3
Data pin
Data pin
Data pin
Data pin
Data pin
Data pin
Data pin
Data pin
2
3
4
5
6
7
8
9
USB / Microprocessor
The FCD4A14 is easy to connect to a microcontroller that
will manage the USB protocol. The same applies for a
microprocessor / microcontroller / DSP.
PCLK
ACLK
RST
OE
Data strobe pin 14
Acknowledge pin 11
pin 16
(see also FC15A140 application note 02).
The USB microcontroller will send the data read on the
data bus directly on the USB cable. A program, called firm-
ware, is run on the USB microcontroller, and another
program, called the driver, is run on the host computer
(generally a PC).
GND
TPE
pin 17
A transfer must be done, with a proper bandwidth reserva-
tion, to make sure that the acquisition is regular without
skipping frames.
TPP
V
CC
USB port
Software
Atmel doesn’t provide specific authentication software with
the FingerChip. Imaging software is provided with the dem-
onstration kit, so that it will be possible to evaluate the
sensor’s capabilities (standard bitmap image files of the fin-
gerprint may be saved), but no matching software for
extracting minutia and performing comparisons is provided.
FCD4A14
USB micro
Do0-3, De0 -3
PCLK
Data
clock
USB
cable
8
FingerChip is compatible with software adapted to optical
sensors, but it may be better to take advantage of the spe-
cific features of the device-particularly the fact that large
images with more information may be obtained, thus
enabling a reduction of the FAR & FRR.
RST
OE
output
output
output
Many of our FingerChip Partners have adapted (or are in
the process of adapting) their algorithms and/or matching
hardware to the FingerChip. If you need more information
on these products, please contact Atmel or visit our web
site.
TPE
TPP
VCC
Reducing Area: Sweeping the Finger Over the
Fingerchip
Reducing the cost of the sensor is one of the most impor-
tant topics in fingerprint capture. In silicon sensors
particularly, the smaller the area, the less expensive the
device.
Parallel Port
Parallel port must conform to the EPP specification for
speed purposes. A standard bi-directional parallel port is
able to acquire only about 200 kilobyte per second (it
depends on the PC): this is three times slower than EPP.
Thus the maximum finger speed is reduced 3 times.
FingerChip technology delivers this size reduction by using
an array with very few columns. A user sweeps his/her fin-
ger over the sensor, and FingerChip delivers a burst of
images.
The FCD4A14 can be directly connected to the parallel port
without interface glue.
FCD4A14
14
FCD4A14
At this time, two strategies are possible:
• reconstruct the complete fingerprint image, and then
perform authentication
The EPP version of the library is delivered with the parallel
port kit.
• analyze the images on the fly, and decide user is
Start And Stop Acquisition
recognized if enough images are matched
When the user is asked to sweep his/her finger, the system
begins to analyze incoming images to detect the finger and
avoid storing blank frames (in the case of storing images
for later reconstruction, if not done on the fly).
Analysis on the fly is better from a user point of view,
because the acceptance can be given before the end of the
finger’s sweep.
The clock rate has to be chosen carefully to enable the
user to move his/her finger quite quickly, while also obtain-
ing many overlapping images. With the FingerChip, it is
possible, using only the images, to reconstruct the com-
plete fingerprint without knowing the real speed of the
finger.
The same problem occurs when acquisition must end. The
provided library contains all the basic routines that perform
these operations. This analysis is done on the fly.
Note that it is very important to have a regular acquisition
without (large) processor interrupts during storage of fin-
gerprint slices (once the finger is detected). If slices are
missing, it may be impossible to reconstruct a complete fin-
gerprint image. Developers must take care to provide a
high priority level during acquisition to the process, so that
heavy applications running at the same time will not
interfere.
Correlating two images to find out the distance between
them seems at first to be too computer-time intensive. Yet
while this is more or less true for the first two images, as we
don’t know the finger’s speed, for the following images, the
position is easily guessed as the speed of the finger is
more or less constant, and a clever strategy may be used.
Ergonomy
Software Library
Special attention must be taken in order to make the sen-
sor easy to use. The second level packaging (the box that
contains the FingerChip and other electronic components)
must allow the user to continuously touch the device during
sweeping. Nothing should be placed in the path of the
finger.
Atmel provides a dynamic linked library (FC_GetImage.dll)
with only one routine to call to get an image. Contact Atmel
for the application note describing all the features of this
library.
This library calls low-level routines to access hardware. Dif-
ferent libraries will exist, depending on the associated
hardware (EPP/USB/etc.), but the call will remain the
same, so developers do not need to update their software if
the hardware changes.
To make sure the finger touches the device, it is better to
allow the user to curl his/her finger around the second level
packaging. This also depends if you are standing or sitting
in front of the device. For instance, it is very difficult to put
your finger flat on a wall in front of you (see figure).
15
In this configuration,
only the tip of the
Tilt the sensor
fingerprint is captured
Prefer these solutions
Put the device on an edge
This part of the fingerprint
is not captured
Or combine an edge and a bump
Preferred solution as your finger
receives sensory feedback
information that indicates where
the sensing area is
Sweeping a finger flat on a wall is a
difficult move when standing up
User can curl
his/her finger
around the sensor
Adjust the position depending
on the situation
keyboard
FCD4A14
16
FCD4A14
Packaging: Mechanical Data
3.15 0.32
25.40 0.25
0.65 0.08
(2.45)
(2.15)
(0.20)
(0.90)
9.36 0.15
6.34 0.15
A
N0.11
N0.20
0.40
N0.1
N0.10
A'
1.00-0/+0.2
0.90
0.80
1.80
0.08
INDEX MARK
4.97
(ø1.50)
CROSS-VIEW A-A'
sensitive area
dam fill (black epoxy)
&
1.1 0.1
0.08
60˚
0.05
0.81
0.05
0.46
2.54 0.13
22.86 0.13
(P=2.54X9)
17
Ordering Information
Package device
FC D4A14 C C —
Atmel prefix
FingerChip family
Quality level
— : standard
Device type
Package
C : DIP Ceramic 20 pins
CB : Chip OnBoard (COB)
Temperature range
Com: 0˚ to +70˚C
FCD4A14
18
Atmel Headquarters
Atmel Operations
Corporate Headquarters
2325 Orchard Parkway
San Jose, CA 95131
TEL (408) 441-0311
FAX (408) 487-2600
Atmel Colorado Springs
1150 E. Cheyenne Mtn. Blvd.
Colorado Springs, CO 80906
TEL (719) 576-3300
FAX (719) 540-1759
Europe
Atmel Rousset
Atmel SarL
Zone Industrielle
13106 Rousset Cedex
France
Route des Arsenaux 41
Casa Postale 80
CH-1705 Fribourg
Switzerland
TEL (33) 4-4253-6000
FAX (33) 4-4253-6001
TEL (41) 26-426-5555
FAX (41) 26-426-5500
Atmel Smart Card ICs
Scottish Enterprise Technology Park
East Kilbride, Scotland G75 0QR
TEL (44) 1355-803-000
Asia
Atmel Asia, Ltd.
Room 1219
FAX (44) 1355-242-743
Chinachem Golden Plaza
77 Mody Road Tsimhatsui
East Kowloon
Atmel Grenoble
Avenue de Rochepleine
BP 123
Hong Kong
38521 Saint-Egreve Cedex
France
TEL (33) 4-7658-3000
FAX (33) 4-7658-3480
TEL (852) 2721-9778
FAX (852) 2722-1369
Japan
Atmel Japan K.K.
9F, Tonetsu Shinkawa Bldg.
1-24-8 Shinkawa
Chuo-ku, Tokyo 104-0033
Japan
TEL (81) 3-3523-3551
FAX (81) 3-3523-7581
Fax-on-Demand
North America:
1-(800) 292-8635
International:
1-(408) 441-0732
e-mail
literature@atmel.com
Web Site
http://www.atmel.com
BBS
1-(408) 436-4309
© Atmel Corporation 2000.
Atmel Corporation makes no warranty for the use of its products, other than those expressly contained in the Company’s standard war-
ranty which is detailed in Atmel’s Terms and Conditions located on the Company’s web site. The Company assumes no responsibility for
any errors which may appear in this document, reserves the right to change devices or specifications detailed herein at any time without
notice, and does not make any commitment to update the information contained herein. No licenses to patents or other intellectual prop-
erty of Atmel are granted by the Company in connection with the sale of Atmel products, expressly or by implication. Atmel’s products are
not authorized for use as critical components in life support devices or systems.
®
™
Marks bearing and/or are registered trademarks and trademarks of Atmel Corporation.
Terms and product names in this document may be trademarks of others.
Printed on recycled paper.
1962A–01/00/xM
相关型号：
©2020 ICPDF网 联系我们和版权申明