We report on the continued development of our 94GHz dual-mode radar/radiometric imager, AVTIS. To date we have concentrated on refining the radar mode and can now acquire state-of-the-art long range, high resolution radar images. More recently we have worked to improve and integrate the radiometric mode, to complete the dual-mode functionality. One notable problem of the monostatic architecture is that of radar transmitter leakage via the circulator/antenna assembly, even with the transmitter signal heavily attenuated. A related issue is the high level of AM noise in the local oscillator signal due to the IMPATT multipliers and power amplifiers used. This far-off-the-carrier noise, in combination with the leakage, desensitises the receiver and raises the noise floor to give a receiver noise temperature of approximately 6000K. This yields a thermal resolution of 2K in 4ms that is considered adequate for the intended volcanological application. In comparison, tests with a high power Schottky diode multiplier chain as the LO source has yielded a noise temperature of 750K in a separate radiometer. Thermal calibration is also of concern and we have implemented an IF noise adding circuit in which the radiometer alternates between the scene and hot and cold IF noise references. This accounts for dominant receiver fluctuations in the IF amplifiers but sky tipping curves are used to calibrate the overall response. Whilst the radar is already capable of producing high resolution topographic maps with surface reflectivity overlaid, it is hoped that co-aligned radiometric brightness temperature data will lead to a better understanding of the emissivity and surface roughness of the terrain being surveyed.

Operation of military helicopters in a dusty environment challenges pilots with reduced visibility. Passive millimeter-wave (MMW) imaging has the potential to be used in these environments to image through dust cloud obscurants. The millimeter-wave phenomenology of the sand environment and the obstacles present in that environment are explored in this work. A 93 GHz polarimetric passive MMW imager was used to characterize an obstacle-rich sand environment and the results are presented. It is shown that there is a strong polarimetric signature present for both sand and cinder block between 10 and 30 degrees depression angles. Also shown is the phenomenology associated with shadows on sand. It was determined that berms and ditches can be very difficult to detect using even a sensitive MMW radiometer. The results can be used to model the performance of passive MMW imaging systems in a sandy environment.

A proto-type passive millimeter-wave (MMW) camera with interferometric processing has been developed. The purpose is to confirm the feasibility of the interferometric MMW camera and to study the characteristics of MMW images. In this paper, the principle and the feature of the interferometric MMW camera is described. Also, the hardware configuration and the image processing algorithm are presented. This proto-type camera is comprised of the minimum configuration as an interferometric imager which consists of two sets of a W-band front end with a horn antenna, a receiver, and an A/D converter, a high-speed processing hardware, and a computer. The position of these two antennas with W-band front-end moves on the precision linear slider in horizontal and vertical axis. The coherently amplified two channel signals are digitized and processed in the hardware processor. The process is comprised of phase error compensation, correlation of all combination of each axis data, and integration to improve the signal to noise ratio. The computer input the integrated data to make an image by matched filter processing. The integration time is from 1mS to 10S depending on required integration gain. The maximum synthesized antenna aperture size is 1m for horizontal axis and 50cm for vertical axis. Because it takes certain time to receive by the moving antennas, only the targets without motion are imaged by this proto-type camera. The processed images will be shown. Also, future plan for a real-time camera using this technique is presented.

As semiconductor costs drop, Millimeterwave (MMW) imaging becomes more attractive to users of remote sensing and security applications. Many cameras have been reported in the millimeter and even Terahertz bands, and early images are always exciting. However, good system models are required to predict operational performance. All bodies above Absolute Zero emit energy in accordance with Planck's law. However, the actual energy emitted depends not only on the object temperature, but the emmisivity and surrounding temperature. Some materials have high emmisivity, such as water or microwave absorber, and they emit primarily their own blackbody energy. Other materials, such as metals, have very low emmisivity, and principally reflect the world around them. While there are no truly exotic physics involved, getting models 'right' in some useful sense is a difficult and time consuming process. A simple ray trace method with a non-polar environment / emmisivity models provides a basis to predict imager performance. An unpolarized illumination source and an unpolarized detector are assumed to begin, which are then extended to include polar effects. Inclusion of polar effects simply requires the complex dielectric constant, and the angle of incidence for the ray being traced.

This paper describes the further development of a scene simulation model to include the phenomenology of scattering. Scattering gives rise to increased absorption within media and forward and backscatter leads to a modification of the subject illumination. The amount of scattering from objects is linked to the relation between the radiation wavelength and the composition of the objects. When the object surface roughness and the particulate composition is of the order of the radiation wavelength, scattering dominates. However, for longer wavelengths, reflections tend to be more specular in character and the wave propagating within the medium retains the spatial coherence. Imaging of people and man made objects in the low frequency band of the millimetre wave region appears to be associated with this longer wavelength regime. However, in moving up in frequency to the sub-millimetre wave band, the fibrous composition of clothing and hair is likely to change this imaging phenomenology. This paper investigates how scattering phenomena can be included in polarimetric scene simulation. Simulated images are shown for passive and active imaging to demonstrate the scattering phenomenology.

A methodology for retrieving land surface properties from passive microwave observations is presented. Dual polarization microwave brightness temperature data, together with a simple radiative transfer model are used to derive surface soil moisture and vegetation optical depth simultaneously, in a non linear optimization procedure using a forward modeling approach. Soil temperature is derived off-line with a common heat flow model, driven by high frequency vertical polarization microwave data and remotely sensed observations of net radiation. The methodology does not require any field observations of soil moisture or canopy biophysical properties for calibration purposes and is independent of wavelength. Remote sensing provides an excellent opportunity to monitor and gather environmental data in regions that have little or no instrumentation. Moreover, microwave technology provides a more all-weather capability than is typically afforded with visible and near infrared wavelengths. The model was developed for regional- to global-scale monitoring and related environmental applications such as surface energy balance modelling, numerical weather prediction, flood and drought forecasting, and climate change studies. However, at higher spatial resolutions, which would be possible with aircraft, especially unmanned vehicles, tactical applications may be realized as well. Retrieval results compare well with field observations of soil moisture and satellite-derived vegetation index data from optical sensors.

Over the past few decades, passive millimeter-wave (PMMW) sensors have emerged as useful implements in transportation and military applications such as autonomous flight-landing system, smart weapons, night- and all weather vision system. As an efficient way to predict the performance of a PMMW sensor and apply it to system, it is required to test in SoftWare-In-the-Loop (SWIL). The PMMW scene simulation is a key component for implementation of this simulator. However, there is no commercial on-the-shelf available to construct the PMMW scene simulation; only there have been a few studies on this technology. We have studied the PMMW scene simulation method to develop the PMMW sensor SWIL simulator. This paper describes the framework of the PMMW scene simulation and the tentative results. The purpose of the PMMW scene simulation is to generate sensor outputs (or image) from a visible image and environmental conditions. We organize it into four parts; material classification mapping, PMMW environmental setting, PMMW scene forming, and millimeter-wave (MMW) sensorworks. The background and the objects in the scene are classified based on properties related with MMW radiation and reflectivity. The environmental setting part calculates the following PMMW phenomenology; atmospheric propagation and emission including sky temperature, weather conditions, and physical temperature. Then, PMMW raw images are formed with surface geometry. Finally, PMMW sensor outputs are generated from PMMW raw images by applying the sensor characteristics such as an aperture size and noise level. Through the simulation process, PMMW phenomenology and sensor characteristics are simulated on the output scene. We have finished the design of framework of the simulator, and are working on implementation in detail. As a tentative result, the flight observation was simulated in specific conditions. After implementation details, we plan to increase the reliability of the simulation by data collecting using actual PMMW sensors. With the reliable PMMW scene simulator, it will be more efficient to apply the PMMW sensor to various applications.

The wideband microwave or millimeter-wave cylindrical imaging technique has been developed at Pacific Northwest National Laboratory (PNNL) for several applications including concealed weapon detection and automated body measurement for apparel fitting. This technique forms a fully-focused, diffraction-limited, three-dimensional image of the person or imaging target by scanning an inward-directed vertical array around the person or imaging target. The array is switched electronically to sequence across the array at high-speed, so that a full 360 degree mechanical scan over the cylindrical aperture can occur in 2-10 seconds. Wideband, coherent reflection data from each antenna position are recorded in a computer and subsequently reconstructed using an FFT-based image reconstruction algorithm developed at PNNL. The cylindrical scanning configuration is designed to optimize the illumination of the target and minimize non-returns due to specular reflection of the illumination away from the array. In this paper, simulated modeling data are used to explore imaging issues that affect the cylindrical imaging technique. Physical optics scattering simulations are used to model realistic returns from curved surfaces to determine the extent to which specular reflection affects the signal return and subsequent image reconstruction from these surfaces. This is a particularly important issue for the body measurement application. Also, an artifact in the imaging technique, referred to as "circular convolution aliasing" is discussed including methods to reduce or eliminate it. Numerous simulated and laboratory measured imaging results are presented.

In this talk we describe the second version of our novel millimeter wave imaging system. This system is an active coherent system, working at 24GHz, and using planar programmable RF-lens. The system is capable of producing realtime movies without processing latency. Moreover, due to the planar nature of the lens, and its compact form, the footprint of the system is very small. The fact that it is monochromatic assures minimal spectral occupancy resulting in an imaging system well suited to personnel screening. The RF-lens, which lies at the heart of the proposed system, is a passive RF element, capable of focusing RF energy into small volumes within its field of view through a mechanism that changes the phase of the impinging wave on the lens-surface. This enables one to scan image voxels (volume pixels) in front of the lens. Since the phase change can be rapidly switched electronically, more than 10^7 voxels are scanned per second. We will discuss the reasons that lead us to develop this second system, the improvements that were made, and the overall enhancements to the system performance. For example, we will discuss the changes that enable us to go to higher frame rate.

It is well known that millimetre waves can pass through clothing. In short range applications such as in the scanning of people for security purposes, operating at W band can be an advantage. The size of the equipment is decreased when compared to operation at Ka band and the equipments have similar performance. In this paper a W band mechanically scanned imager designed for imaging weapons and contraband hidden under clothing is discussed. This imager is based on a modified folded conical scan technology previously reported. In this design an additional optical element is added to give a Cassegrain configuration in image space. This increases the effective focal length and enables improved sampling of the image and provides more space for the receivers. This imager is constructed from low cost materials such as polystyrene, polythene and printed circuit board materials. The trade off between image spatial resolution and thermal sensitivity is discussed.

Passive millimetre wave imaging is now an established and accepted technology that is finding viable commercial applications in many areas, particularly security and border control. The upper frequency of operation has largely been governed by the availability of solid state uncooled detectors to around 100GHz. Passive operation at higher frequencies potentially offers some unique features such as higher optical resolution for a given system size, increased depth of focus and improved scene contrast. However, the technological challenges involved in realising arrays of terahertz detectors with the required sensitivity, packing density, repeatability and reliability are considerable. Nevertheless we have developed such detector arrays and these now provide the core technology base upon which to explore many new commercial applications that require such benefits. The first commercial application for this technology has been in the realisation of a compact passive real time imaging system primarily aimed at the detection of concealed contraband and firearms at a remote distance. This paper describes such an imager. Some of the practical advantages of imagery at these wavelengths, such as the ability to operate un-illuminated either outdoors or indoors, will be described as well as some of the practical system aspects such as bandwidth, scanning methodology and optical design. Examples of typical real time imagery is provided.

Microwaves can be used to detect hidden objects behind optically opaque materials. Hence, the penetration capability through such materials is of fundamental importance. In order to characterise a material of interest in the microwave region, its permittivity should be known besides its physical structure. In many cases the permittivity is unknown, inaccurately known, or known for only specific frequencies. Also very often the range of values given in the literature can have a large variability for a specific situation. In this paper we describe a procedure to determine the permittivity from radiometric free-space measurements of nearly arbitrary materials. The advantage of this method is that large material samples like brick or wooden plates, and materials like textiles, which are hard to mount in a defined way in a waveguide, can be investigated. Some representative results for MMW measurements are shown, and an estimation of the presently achieved precision is given. The first attempts showed a satisfying performance, although not for all materials and frequencies a unique solution could be found.

To obtain high-angular-resolution (< 1 milliradian) passive millimeter-wave images using traditional imaging techniques requires prohibitively large apertures for most applications. Sparse aperture techniques present a viable alternative by providing large effective aperture without the corresponding cubic increase in imager volume and weight. However, implementing a large field-of-view, high-resolution, sparse-aperture imager presents a number of challenges, namely: (1) the fabrication of a large number of phase-sensitive, ultra-low-noise millimeter-wave detectors, (2) routing and (3) realtime correlation of acquired data across the array, and (4) implementation of true time delays for steering the field-of-view without pronounced fringe-washing effects. In previous work, we have presented optical upconversion techniques as a viable alternative for overcoming these challenges. Converting passive millimeter-wave energy collected at each point in the array into sidebands on a common optical carrier signal allows for (1) recording of broadband complex field values onto the optical carrier, (2) efficient routing of these field values via fiber optics, (3) correlation of the recorded data via optical filtering and simple optics, and (4) implementation of true time delays using fiber optic or integrated optical switches. Herein, we present continued theoretical and experimental progress toward the realization of such optically based sparse arrays for passive millimeter-wave imaging applications.

The Ultra Sensitive Silicon Sensor¹ is an all silicon bolometer that utilizing active thermal isolation to overcome limitations in conventional bolometers. Inherently, bolometers are slower and less sensitive than quantum detectors. However, because of low cost and room temperature operation, bolometers are being used in many cameras. IC technology used for producing staring focal planes has ameliorated the bolometer's slow speed and lower sensitivity issues. Better thermal isolation and thermal responsivity materials have yielded significant sensitivity improvements. Currently, LWIR bolometer's sensitivity is about 10X below the theoretical limit. Achieving theoretical sensitivity will greatly increase the application sphere of bolometer cameras. Theoretically, performance improvements are possible with better thermal isolation, better thermal responsivity, and a faster time constant. Examining these limitations has allowed us to formulate solutions to these problems and open the opportunity for application of bolometers to MMwave imaging. Large (10X) sensitivity improvements are possible by replacing passive thermal isolation with active thermal isolation. Active thermal isolation utilizes electro-thermal feedback to greatly improve thermal isolation and this leads directly to corresponding responsivity improvements. Improved thermal isolation does increase the thermal time constant, however, this increase is offset by using microantennas and/or microlenses. A detail analysis is presented on the theoretical operation of the all silicon USSS bolometer.

We have demonstrated a high efficiency package for zero bias Sb-based backward tunnel diodes developed for passive millimeter wave imaging. Flip-chip mounting of detector MMICs onto quartz substrates permit placement of the detector directly within the WR-10 waveguide feeds for diagonal horn antennas. This arrangement minimizes the losses between the detectors and antennas while providing an impedance match over a majority of W band. A 2x2 array of radiometers was fabricated, assembled, and measured using coherent measurement techniques. The resulting noise equivalent temperature difference, calculated assuming a 30 Hz frame rate is 10 degrees K.

Sb-heterostructure diodes have become the detector of choice for W-band millimeter wave imaging cameras now being commercialized or in prototype development. Here we optimize the diode impedance to yield a minimum noise equivalent power (NEP). The goal is to decrease the gain required of the front-end LNA. Measured W-band sensitivities for two diodes are 3500 and 5500V/W. Their zero bias differential resistance values imply Johnson noise limited NEP's of 0.98 and 0.83pW/Hz^1/2, respectively, much less than obtained from conventional biased Schottky diodes. A MMIC version of the diode detector has been simulated with an integrated bandwidth of ~ 30 GHz at W-band. The simulated temperature sensitivity (NEΔT) with an HRL W-band LNA on the front end is <1°K.

If the object is a point whose presentation being the delta-function d, then we have at the output the receiving system: Point Spread Function (PSF) O=O*d, asterisk denoting a convolution. We have developed the ultra-resolution method. It enables us to solve the problem of the inversion for distortion of PSF R*O~d, where R is resolving function [1, 2]. Let T be a image of the object-target. Then its image at the output of the receiving system is O*T. The following problem is set forth: it is necessary to construct a function R such that should be valid the following statement: R*(O*T) ~d(r-r0). The obtained function R enables us to convert the receiving image O*T into delta-function d(r-r0) which refers to the definite point r=r0 of the object-target T [3]. The concept of characteristic is introduced in order to optimize object-targeting system. The method may be embedded in the active radar or passive multi-rays radio-vision systems for resolving and for point indication of object-targets. There are numerous examples of the solutions to the model problems.