Citation Link: https://nbn-resolving.org/urn:nbn:de:hbz:467-950
Bestimmung und Modellierung von Detektionsgrenzen bei TFA-Bildsensoren
Source Type
Doctoral Thesis
Author
Subjects
TFA-technology
Image sensors
Fixed-pattern-noise
DDC
620 Ingenieurwissenschaften und Maschinenbau
GHBS-Clases
Issue Date
2005
Abstract
Thin-Film-on-ASIC (TFA) sensors use a thin film layer as photoelectric detector. Based on
amorphous silicon, the film is deposited on top of the 2-dimensional crystalline pixel area.
This thesis analyses the limiting properties of integrating TFA image sensors, i.e. sensitivity,
dark current, temporal and fixed pattern noise.
A useful current sensitivity is obtained with optimized detector geometries. Parasitic capacitances
of the 3-dimensional design and the selected pixel input circuit determine the
conversion or voltage sensitivity.
The dark current is determined by thermal effects or electrical field effects. The reverse
current of the detector, the leakage current of the reset transistor and the leakage current of the
driver gate have to be considered. The dark current temperature dependence varies with the
dominating physical property. Dark current improvements of ohmic components by cooling
are negligible.
The integration of photo and dark current detector noise into SPICE enables the determination
of the signal-to-noise ratio and the dynamic range of the pixel input stage. In most cases, the
driver transistor noise determines the detection limit.
A description of gain FPN (PRNU) depending on the transistor geometry is developed using
the moment method. Temporal noise has a stronger influence on the pixel input stage than
fixed pattern noise.
Due to the generally valid description, the central results of this thesis are applicable to
normal CMOS as well as TFA image sensors.
amorphous silicon, the film is deposited on top of the 2-dimensional crystalline pixel area.
This thesis analyses the limiting properties of integrating TFA image sensors, i.e. sensitivity,
dark current, temporal and fixed pattern noise.
A useful current sensitivity is obtained with optimized detector geometries. Parasitic capacitances
of the 3-dimensional design and the selected pixel input circuit determine the
conversion or voltage sensitivity.
The dark current is determined by thermal effects or electrical field effects. The reverse
current of the detector, the leakage current of the reset transistor and the leakage current of the
driver gate have to be considered. The dark current temperature dependence varies with the
dominating physical property. Dark current improvements of ohmic components by cooling
are negligible.
The integration of photo and dark current detector noise into SPICE enables the determination
of the signal-to-noise ratio and the dynamic range of the pixel input stage. In most cases, the
driver transistor noise determines the detection limit.
A description of gain FPN (PRNU) depending on the transistor geometry is developed using
the moment method. Temporal noise has a stronger influence on the pixel input stage than
fixed pattern noise.
Due to the generally valid description, the central results of this thesis are applicable to
normal CMOS as well as TFA image sensors.
File(s)![Thumbnail Image]()
Loading...
Name
sterzel.pdf
Size
2.35 MB
Format
Adobe PDF
Checksum
(MD5):0cb1bb0ea4dba28eb9926712e45b8e51
Owning collection