Design approaches to tactical air defense are outlined. A SAM and an air-to-air missile trajectory are used to optimize a system design for maximum target detection range, taking system noise and background clutter noise into account. Detection ranges of 30 km are found to be feasible except for bearing angles close to 90 and 270 deg when the background is between 35 and 75 km distant. Methods to operate in these gaps are suggested. For a 2D Dynamic Stare simulation using a real earth background, background noise was cut by five orders of magnitude, representing a base case type of performance for Dynamic Stare. Changes to make the system high resolution and high sensitivity are discussed.
The usefulness of Z-plane technology to GEO instrumentation with regard to data fidelity, data compaction, and parallel processing is discussed. The application of Z-plane technology to imaging spectrometers and advanced lightning mappers is addressed. Desirable design characteristics for Z-plane signal processing are given.
Space-based sensors currently transmit large amounts of raw data to their receiving stations, forcing
large bandwidths to be used and large data archives maintained. One possible solution to this data
management problem is the use of more intelligent processing onboard the spacecraft. Neural networks
are proposed as a trainable, mutable means of achieving sensor signal processing, sorting, and
classification before infrared (IR) focal plane data leaves the cryogenic seal. Two questions arise: (1)
can a neural network of appropriate size learn a sensor processing classification problem, and (2) will
there be devices to implement that neural network as a smart focal plane. In this paper we concentrate on
the first issue and show successful results in simulation for a massively parallel, distributed neural
network solution to the closely spaced object (CSO) recognition problem, using a defocused JR sensor
model with noise, and an uncooperative object or CSO at large distance. The back-propagation learning
method is used to train the network. Architectural issues regarding the construction of a prototype
neuromorphic focal plane device are also discussed.
Z-plane technology affords the implementation of powerful image processing algorithms which operate with very high
speed and with very low power consumption. This paper discusses architectures for implementing conventional image
processing operations such as edge detection and enhancement, feature extraction, image encoding, and transform computation.
Many of the implementations presented assume the use of cascaded orthogonal Z-Plane modules.
Neural networks offer the potential for a quantum leap in the capabilities of imaging sensor systems. The critical
neural network implementation factors are: weighted interconnect between all detector outputs; parallel, linear processing
of each detector output; fan-out to multiple (thousands) processing nodes per detector output and the ability to independently
change interconnect weights and processor node connections within the detector integration times. For a 128 x 128 pixel
detector array, the number of desirable interconnects could be as high as iO per second, compared to the approximately
iO rates achieved presently with off-focal plane digital processors. Irvine Sensors Corporation (ISC) has conceived a new
way of interconnecting 3-D focal plane readout modules and of laying out their component integrated circuits that appears
to fulfill the very high interconnect rate requirements. This concept is described and mterconnectivity and other performance
attributes are discussed.
This paper describes the key technology, onboard processing, and integration of a space reconnaissance platform consisting of a small smart satellite (Smartsat) designed for compatibility with the launch envelope of a Pegasus class vehicle. Because the platform is designed to the Pegasus class baseline, the operations scenario can use a launch-on-demand concept that provides inherent advantages for tactical force commanders that are not achievable in today's environment. In a very significant departure from traditional designs, Smartsat houses sufficient onboard digital processing so that usable data can be downlinked directly to the forward-located force commanders. This feature minimizes the requirement for high bandwidth data links to the ground and the consequent dependence on large, vulnerable central processing ground stations to produce and distribute usable data for the decision makers. An example of how Smartsat might be applied to tactical air defense is given.
Miss distance performance has generally required a significant simulation effort. Missile simulation
models generally require substantial modification to adapt them to a new system concept or to make
major changes within the interceptor design. By conducting a sensitivity analysis, an alternate approach
is presented to assist the analysis of the the miss and establishing the system level requirements.
The analysis section of this paper analyzes each of the three miss components aimpoint determination,
measurement accuracy, and interceptor dynamic response. Observation and sample calculations are
presented for a maneuvering and a non-maneuvering target. Once the miss requirement has been
analytically determined as a function P, then the analytical process can be worked backwards to
determine the range and range rate uncertainty requirement at handover, or if at the start of end
game engagement, the total angular accuracy (in a r.s.s sense) of the system and required data rate
may be determined.
Extremely low power superconductive electronics (SCE) (low noise preamplifier, analog to digital converter, multiplexer,
etc.) for very large focal plane arrays can significantly reduce the overall sensor system power, hence its weight and volume,
thus reducing overall mission cost. The general architecture for a Z-plane, all-superconducting technology focal plane signal
processor is presented illustrating the functional elements and their general configurations. The low noise and speed of the
TRW developed SCE permits unique solutions to focal plane array signal processing issues such as in-line gamma
suppression and digital signal integration.
This paper relates spike-adaptive TDI performance to the level of the background encountered and describes alternative approaches to spike-spike suppression that void the baseline equalization problem. The 'Lowest of N' and the 'Iterative Lowest of N' algorithms are identified as spike suppression methods that are independent of detector baselines.
This paper describes the design and testing of a low power Analog to Digital
converter. In the design of Z-Plane focal plane array technology the
consumption of power by circuitry in the signal processing electronics, that
are part of the Z-Plane, is a primary limiting factor in the overall Z-Plane
system signal processing architecture. The Analog to Digital converter was
designed by applying charge-coupled device (CCD) technology to the binary
weighing problem. The Analog to Digital converter is, with the exception of
the comparator, an all digital CMOS design. The design concepts are discussed
along with preliminary test results.
Algorithms are described for ordering, averaging, and spike adaptive non-linear
gamma circumvention. These algorithms take advantage of the nominally matched-insignal
and position-correlated samples from the detectors of the time delay integration
(TDI) chain of a scanning sensor. The effects of averaging and median (ordering)
filtering on Gaussian noise and Gaussian plus gamma noise are presented along
with the requirement for additional post filtering for these algorithms. Monte
Carlo techniques are used to predict the performance of the various circumvention
algorithms for both endoatmospheric and exoatmospheric applications. Finally,
selected gamma circumvention algorithms suitable for on-focal plane mechanization
using large scale integration (LSI) technology are presented.
This paper provides an overview of the current state-of-the-art HYMOSS (Hybrid Mosaic On Stacked Silicon) Z-technology. In the first part of this paper, an introduction to the HYMOSS physical characteristics is presented. This includes a description of the stacked substrates (cube) and mounting hardware (module). The basic steps in manufacturing HYMOSS are covered. The paper concludes with a description of the two newest endeavors for HYMOSS technology: stacking of superconducting ICs, digital memory, and processor ICs.
The process for manufacturing parallel processors using Z-plane technology is described along with the interconnectivity achievable for parallel processors. A thermal analysis performed on a typical module using the NASA SINDA model is shown. The Z-plane packaging technology producibility and reworkability are addressed.
The transition from the relatively costly, low-rate pilot production of Z-technology focal planes to high-rate production in a cost-effective environment is discussed. The use of computer-integrated manufacturing and automation in this transition is addressed. The proposal process and systems specification in the transition is reviewed.
This paper addresses the challenges in creating and developing an automated factory for the production of mosaic focal plane arrays. It is shown how the modular concept has been employed to its maximum benefit in the development of the array. Finally, it is demonstrated how solid strategic planning can result in significant benefits beyond the immediate solution for the problem at hand; that is, turning a point solution into an industry solution.
Quality assurance, wafer lot production, device testing, postprobe processing, and qualification and quality screening of tape automated bonded devices are discussed. These devices are qualifed for space applications and ready to assemble directly into production modules.
Tape automated bonding processes have been adapted to the development of a fully automated cell. Details of the early cell design and planning are presented, including a discussion of reliability and producibility advantages achievable from the cell and some early problems encountered in the development. Test results from the first demonstration modules are presented along with projections for results considered achievable with the automated cell.
Performance testing of large area mosaic focal planes requires high speed and high
volums test methods because of the large number of individual elemsnts to be tested.
Extensive use of parallel activities is required to avoid the sequential delays
involved in loading, evacuating, cooling, testing, warming, pressurizing, and
unloading the test articles. A parallel path production test system designed for
automated testing of up to twelve thousand detector channels per hour will be
In a highly automated, high rate, "hands off" production system, in-process
inspections can no longer be isolated from the production processes, but must be commingled chronologically and physically within the manufacturing processes and their
applicable equipments. This must be accomplished, however, without losing the
autonomy and purposes of inspection in the classical sense. This paper will not only
show how inspections can be maintained in such an environment of automation, but how
processes, yields and sample inspection levels can be enhanced by real time and
near-term correlation of process and inspection variables.
The fabrication of the first Z-technology module with 128 active layers of readout electronic circuitry is being pursued. Several issues which will be investigated regarding the 128-layer module build, such as photolithographic processing at the edge of the module (edge effects) and module buttability, are discussed in this paper. Other topics which are covered apply to future applications of fully populated HYMOSS modules. These topics include possible reductions of thermal mass for quick cool-down time applications and improvements of end-to-end yield for low-cost applications.
This paper examines three applications of automation technology in the manufacture of Infrared Focal Plane arrays. Areas to
be examined ar wafer handling during array fabrication up through dicing, automation ofLiquid Phase Epitaxy, and automation
of bump bonding. The collection of factory data and its use in control of the manufacturing process will also be discussed. Specifics
related to the production of Z-module architecture will be emphasized.
Issues concerning IR system design that drive Z-plane technology development are discussed. Key characteristics of IR phenomenology are briefly addressed, followed by a discussion of critical IR focal plane array performance parameters as they relate to sensor design tradeoff issues. Possible future directions for radical approaches to Z-plane IR focal plane arrays are suggested.