Rochester Institute of Technology (RIT) and its collaborators at the University of Rochester and Harris Corporation are developing a room-temperature imaging Terahertz (THz) frequency detector using Si-MOSFET (Silicon Metal Oxide Semiconductor Field Effect Transistor) CMOS devices. They are implemented into a focal plane imaging array for use in many applications, such as transmission or penetration imaging and spectroscopy. Technology for THz detection is often extremely costly, due to either expensive detector materials or cryogenic cooling systems. However, the devices tested here are low-cost due to the use of conventional room temperature silicon CMOS technology. The devices operate from 170 to 250 GHz with an additional detector design has been fabricated for 30 THz (10 microns wavelength). Results are presented for the initial testing of single test structure FETs. These devices were designed with several different antenna configurations and a range of MOSFET design variations for evaluation. The primary goal of the work presented here is to determine the optimized detector design for the subsequent focal plane array implementation based on the largest responsivities and lowest noise-equivalent power (NEP). Transmission testing of the devices yields responsivities of about 100 to 1000 V/W and a NEP of about 0.5 to 10 nW·Hz<sup>-1/2</sup>. Through this evaluation and by utilizing signal amplification on the chip, signal modulation at higher frequencies, and smaller process sizes the performance of these devices will continue to improve in future designs.
Exelis Geospatial Systems and its CEIS partners at the University of Rochester and Rochester Institute of Technology are developing an active THz imaging system for use in standoff detection, molecular spectroscopy and penetration imaging. The current activity is focused on developing a precision instrument for the detection of radiation centered on atmospheric windows between 200 GHz and 400 GHz (available sources). A transmission imager is developed by raster scanning through a semi-coherent non-ionizing beam, where the beam is incident on a NMOS FET detector. The primary goal of the initial system is to produce a setup capable of measuring responsivity and sensitivity of the detector. The Instrumentation covers the electromagnetic spectral range between 188 GHz and 7.0 THz. Transmission measurements are collected at 188 GHz in order to verify image formation, responsivity and sensitivity as well as demonstrate the active imager’s ability to make penetration images.
Collaboration between Exelis Geospatial Systems with University of Rochester and Rochester Institute of Technology aims to develop an active THz imaging focal plane array utilizing 0.35um CMOS MOSFET technique. An appropriate antenna is needed to couple incident THz radiation to the detector which is much smaller than the wavelength of interest. This paper simply summarizes our work on modeling the optical characteristics of bowtie antennae to optimize the design for detection of radiation centered on the atmospheric window at 215GHz. The simulations make use of the finite difference time domain method, calculating the transmission/absorption responses of the antenna-coupled detector.
Interest in array based imaging of terahertz energy (T-Rays) has gained traction lately, specifically using a CMOS process due to its ease of manufacturability and the use of MOSFETs as a detection mechanism. Incident terahertz radiation on to the gate channel region of a MOSFET can be related to plasmonic response waves which change the electron density and potential across the channel. The 0.35 μm silicon CMOS MOSFETs tested in this work contain varying structures, providing a range of detectors to analyze. Included are individual test transistors for which various operating parameters and modes are studied and results presented. A focus on single transistor-antenna testing provides a path for discovering the most efficient combination for coupling 0.2 THz band energy. An evaluation of fabricated terahertz band test detection MOSFETs is conducted. Sensitivity analysis and responsivity are described, in parallel with theoretical expectations of the plasmonic response in room temperature conditions. A maximum responsivity of 40 000 V/W and corresponding NEP of 10 pW/Hz<sup>1/2</sup> (±10% uncertainty) is achieved.
Exelis Geospatial Systems and its CEIS partners at the University of Rochester and Rochester Institute of Technology
are developing an active THz imaging focal plane for use in standoff detection, molecular spectroscopy and penetration
imaging. This activity is focused on the detection of radiation centered on the atmospheric window at 215.5 GHz. The
pixel consists of a direct coupled bowtie antenna utilizing a 0.35 μm CMOS technology MOSFET, where the plasmonic
effect is the principle method of detection. With an active THz illumination source such as a Gunn diode, a design of
catadioptric optical system is presented to achieve a resolution of 3.0 mm at a standoff distance of 1.0 m. The primary
value of the initial system development is to predict the optical performance of a THz focal plane for active imaging and
to study the interaction of THz radiation with various materials.
We describe preliminary design, modeling and test results for the development of a monolithic, high pixel density,
THz band focal plane array (FPA) fabricated in a commercial CMOS process. Each pixel unit cell contains multiple
individual THz band antennae that are coupled to independent amplifiers. The amplified signals are summed either
coherently or incoherently to improve detection (SNR). The sensor is designed to operate at room temperature using
passive or active illumination. In addition to the THz detector, a secondary array of Visible or SWIR context
imaging pixels are interposed in the same area matrix. Multiple VIS/SWIR context pixels can be fabricated within
the THz pixel unit cell. This provides simultaneous, registered context imagery and "Pan sharpening" MTF
enhancement for the THz image. The compact THz imaging system maximizes the utility of a ~ 300 μm x 300 μm
pixel area associated with the optical resolution spot size for a THz imaging system operating at a nominal ~ 1.0
THz spectral frequency. RF modeling is used to parameterize the antenna array design for optimal response at the
THz frequencies of interest. The quarter-wave strip balanced bow-tie antennae are optimized based on the
semiconductor fabrication technology thin-film characteristics and the CMOS detector input impedance. RF SPICE
models enhanced for THz frequencies are used to evaluate the predicted CMOS detector performance and optimal
unit cell design architecture. The models are validated through testing of existing CMOS ROICs with calibrated THz
Using band-limited LASER speckle to measure the Modulation Transfer Function (MTF) of an image
sensor offers simplified procedure and inexpensive laboratory set up compared with the traditional method
of using a knife edge on the sensor imaging plane. This speckle technique has been previously
demonstrated by Glen Boreman's group on devices in the visible range. We have extended the procedure
to short-wave infrared (IR) sensor at 1.55 micron. Similar measurements were also made at 532 nanometer
on a commercial visible (VIS) sensor. The experiments show that the LASER speckle method to be
accurate when compared to knife-edge measurements for data below Nyquist. The measured MTF data
support optical system design and image quality modeling for both VIS and IR sensing applications.
Kalman filters have been used as a robust method for object location prediction in various tracking algorithms for
nearly a decade. More recently, adaptive and extended Kalman filters have been employed, making predictions
even more reliable. The presented addition to this trend is the employment of a polynomial fit to the history of
object locations, using the adaptive Kalman filter framework. This allows the linear state model of the adaptive
Kalman filter to predict non-linear motion, making tracking more robust. This modified filter will be used in
conjunction with the Mean Shift algorithm as the measurement step. Another important consideration when
using a Kalman filter in this manner will be which correlation coefficient is used. The Pearson product-moment
correlation coefficient is shown to provide more robust tracking when compared to the Bhattacharyya coefficient
when objects have either low resolution or are unresolved.
The development and testing of thermal signature tracking algorithms burdens the developer with a method o f testing the
algorith m's fidelity. Collected video is normally used for testing tracking algorithms to evaluate performance in a variety
of configurations. Acquiring suitable volumes of collected video data in multiple configurations can be prohibitive. As
an alternative to collected video, the development of accurate synthetic thermal infrared vehicle models are incorporated
into background infrared scenes generated using the Digital Image and Remote Sensing Image Generat ion (DIRSIG)
software package. Additional software models for thermally emissive targets and motion are being implemented. The
goals are to accurately incorporate thermal signatures of moving targets into realistic radiomet rically calibrated scenes.
This aids in evaluating tracking algorithms using both visible and thermal infrared signatures for improved day and night
detection capability. The software packages are integrated together to produce synthetic video.
ITT Geospatial Systems has space-qualified a visible band interline Charge Coupled Device (CCD) image
sensor with 18 million pixels developed using commercial technology. The sensor is comprised of an 4320
(H) x 4144 (V) array of 8 micron square pixels. With multiple analog outputs each operating at 20 MHz
the sensor will support 30 frames per second continuous video capture. The pixel incorporates a pinned
photodiode, vertical overflow drain and microlens to achieve low dark current, lag-free imaging with highspeed
global electronic shutter at high quantum efficiency (QE). The vertical and horizontal CCD's are
true two-phase designs which support an integrate-while-read operation. The sensor chip is mounted on an
Aluminum Nitride co-fired ceramic package optimized for electrical signal integrity, thermal and optical
stability. The architecture supports quadrant redundancy. The complete assembly has been space-qualified
to a Technology Readiness Level (TRL) of 6 with Total Ionization Dose (TID) radiation testing at 25 Krad.
The sensor exceeds 12-bit of dynamic range and 31% QE with 5 W of total power. The nonlinearity is
measured to be 1.0% while the global non-uniformity is less than 2%. The low defect density of the CCD
sensor allows high resolution video imaging in a space environment.
We describe a compact, multi-sensor design architecture capable of providing both spectral-polarimetric imaging and
adaptive matched filter target detection in real-time. The sensor suite supports airborne broad-area search missions using
multiple large-format, high speed TDI scanning sensors. The technology approach leverages Micro-Electro-Mechanical
System (MEMS) based spectral imaging systems and scanning TDI arrays originally developed for space based remote
sensing. The MEMS spectrometer system can dynamically select and switch linear combinations of single or multiple
VNIR/SWIR spectral bands with 5nm sampling resolution using a programmable MEMS mirror. The MEMS spectral
filter is capable of providing high quality spectral filtering across a large format sensor with > 1MHz optical switching &
update speeds. A dual instrument sensor suite architecture called the "PRISM sensor" has been developed which is based
on this technology and provides simultaneous spectral-polarimetric imaging and matched filter target tracking with
minimal on-board computing requirements. We describe how this technology can simultaneously perform broad-area
imaging and target identification in near real-time with a simple threshold operation. Preliminary results are illustrated
as additional layer of target-discriminate geospatial information that may be fused with geo-referenced imagery.
The development and testing of thermal signature tracking algorithms burdens the developer with a method
of testing the algorithm's fidelity. Although actual video is normally used for testing tracking algorithms, to
evaluate performance in a variety of configurations, the acquisition of suitable video data volume is
prohibitive. As an alternative to actual video we are developing accurate synthetic thermal infrared models
of vehicles that will be incorporated into background infrared images generated using the Digital Image and
Remote Sensing Image Generation (DIRSIG) software package. Motion for the targets within the
background scene is generated using the open-source Simulation of Urban MObility (SUMO<sup>TM</sup>) software
package. ThermoAnalytics' Multi-Service Electro-optic Signature (MuSES<sup>TM</sup>) software package is used to
model thermal emission from the object of interest. The goal is to accurately incorporate thermal signatures
of moving targets into realistic radiometrically calibrated scenes, and to then test and evaluate tracking
algorithms using both visible and thermal infrared signatures for improved day and night detection
capability. The software packages have been integrated together for a synthetic video
ITT Industries Space Systems Division and Eastman Kodak Company have developed a scalable, data- and power-efficient imaging spectrometer system with a digitally tunable optical filter capability, which enables the rapid selection of high-quality user-defined optical spectral band(s) of interest. The system utilizes a custom-designed, high-contrast diffractive MEMS device with 50 independent spectral switches at the image plane of a double-pass dispersive/de-dispersive
spectrometer. The custom MEMS device is based on grating electromechanical system (GEMS) display technology, which provides very high image contrast (2000:1), fast optical switching speeds (< 100 ns), and a large active area with a very high fill factor. The system enables the selection of arbitrary, narrow or wide spectral bands of interest across the visible spectrum with a sampling resolution of 5 nm, without any moving mechanical parts. The
resulting optical filter quality and performance is comparable to conventional fixed-band dichroic filters used in current remote sensing systems. The brassboard systems are designed for rapid transition to space-based, electro-optical (EO) remote sensing missions that utilize large format linear TDI scanning sensors and large format area staring arrays in the visible band. This technology addresses numerous capabilities to meet future EO system requirements for rapidly selecting and utilizing a high quality imaging optical bandpass of interest. The system concept provides capability for a
>20X scan rate advantage over conventional hyperspectral imagers as a result of the compatibility with TDI scanning. The image quality is comparable to current MSI and HSI systems.