Publications

• The First "Switchboard In the Sky": An Alltonomous Satellite-Based Access/Resource Controller
• The Synchronous Processor
• Microchannel Heat Sinks
• Characteristics of Microbursts in the Continental United States
• Space Surveillance With Medium-Wave Infrared Sensors
• Microwave and Millimeter-Wave Resonant-Tunneling Devices
• Neural Network Classifiers for Speech Recognition

The First "Switchboard In the Sky": An Alltonomous Satellite-Based Access/Resource Controller
Marilyn D. Semprucci

Satellite communications systems can connect hundreds of dispersed users instantaneously. But when such a system is serving many small, widely dispersed users and is handling fluctuating traffic loads, the satellite channels must be allocated dynamically. A satellite-based access/resource controller, a "switchboard in the sky," has long been seen as the most efficient way to use the costly resources of a satellite. Two satellite communications packages, which will provide highly protected links to small mobile terminals, incorporate the first autonomous satellite-based switchboard.

The Synchronous Processor
Ira H. Gilbert

The Synchronous Processor is a single-instruction, multiple-data-stream, special-purpose computer, which can perform calculations at a rate of almost 400 million instructions per second (MIPS). To develop this computer, a new hardware design was implemented and a special programming language was constructed.

Microchannel Heat Sinks
Richard J. Phillips

Microchannel heat sinks can be used in a wide variety of applications, including microelectronics, diode laser arrays, and high-energy-laser mirrors. Heat sinks that can be used to cool diode laser arrays have been fabricated in indium phosphide (lnP) with a thermal resistance as low as 0.072°C/(W/cm²), which allows these devices to dissipate loads in excess of 1,000 W/cm². This thermal resistance is nearly two orders of magnitude lower than that achieved by the methods presently used in the microelectronics industry. A heat-sink thermal- and fluid-performance model is presented; microchannel fabrication techniques are described for InP and aluminum.

Characteristics of Microbursts in the Continental United States
Marilyn M. Wolfson

Microbursts—powerful downdrafts generally associated with thunderstorms that occur in hot, humid weather—have caused a number of aircraft crashes. To prevent future accidents, air traffic controllers must be able to detect, and predict, microburst events. All microbursts are not alike, however; several distinct weather patterns can produce microbursts. Thus a categorization of the different types of microbursts is an essential part of understanding these hazardous phenomena. Using this categorization, the relative hazard to aviation of the various types of microbursts can be assessed.

Space Surveillance With Medium-Wave Infrared Sensors
Michael J. Cantella

To make the coverage of the rapidly expanding satellite population more extensive and timely, current operational space surveillance systems must be upgraded. Infrared sensors can see satellites in the earth's shadow, thereby enabling these sensors to supplement the capabilities of visible-spectrum devices. Sensors that operate in the medium-wave infrared region have good sensitivity and resolution; signal degradation due to the earth's atmosphere is minimal. A practical sensor implementation for space surveillance consists of a large staring array of Si Schottky-barrier detectors. Both ground-based and satellite-borne systems will provide excellent sensitivity and coverage.

Microwave and Millimeter-Wave Resonant-Tunneling Devices
T.C.L.G. Sollner, E.R. Brown, and H.Q. Le

Resonant-tunneling devices, which may be capable of operation at terahertz frequencies, have been developed and tested. Included in these solid state microelectronic components are oscillators, self-oscillating mixers, and harmonic multipliers. A characteristic of these devices, negative differential resistance (NDR), has been observed at room temperature. Resonant-tunneling transistors, which promise operation in the terahertz frequency range, are also proposed.

Neural Network Classifiers for Speech Recognition
Richard Lippmann

Neural nets offer an approach to computation that mimics biological nervous systems. Algorithms based on neural nets have been proposed to address speech recognition tasks, which humans perform with little apparent effort. In this paper, neural net classifiers are described and compared with conventional classification algorithms. Perceptron classifiers trained with a new algorithm, called back propagation, were tested and found to perform roughly as well as conventional classifiers on digit and vowel classification tasks. A new net architecture, called a Viterbi net, which recognizes time-varying input patterns, provided an accuracy of better than 99% on a large speech database. Perceptrons and another neural net, the feature map, were implemented in a very large-scale integration (VLSI) device.