Search

The design, simulation, manufacture and measurement of a W-band five-fold (5F) helically corrugated waveguide (HCW) is reported. The 5F HCW is based on the coupling of the traveling TE31 and near cut-off TE22 modes to create an operating eigenwave. The fabricated test structure has circular waveguide ports and features elliptical polariser sections and broadband TE11 to TE31 mode converters on either side of the 5F HCW. The optimised mode converter design, based on a four-fold (4F) HCW, has a predicted power conversion efficiency greater than 90% from 89 to 102.5 GHz, and 96% peak efficiency at 94 GHz. The optimization of the 5F HCW geometry produced an eigenwave suitable for gyro-devices, but the optimization could equally well have been directed to applications such as pulse compression and microwave undulators. Analysis of simulated electric field profiles showed that the propagating power in the 5F HCW was increased by a factor of 6 over that in the 3F HCW at equivalent peak electric field strength. This is due to the larger diameter of the waveguide. Test structures were manufactured through a combination of precision machining of a sacrificial mandrel, copper growth by electroforming followed by removal of the aluminium mandrel by chemically etching. Measurements of the 5F HCW structure’s dispersion showed excellent agreement with the prediction over the design range of 90 to 98 GHz.

Fast aluminum-gallium-nitride–based metal–semiconductor–metal ultraviolet photodiodes were successfully designed, fabricated, and characterized using conventional photolithography techniques. Various electrode geometries were fabricated to investigate the influence of metal contact shapes on the device performance indices with emphasis on the response speed and bias-voltage–independent efficiency. Based on the independently measured Hall mobility and electric field, we evaluated the carrier transit time of the devices as 1.31 ps with bias-voltage–independent external quantum efficiency of 1198% in photoconductive mode at 19.5 V and 70% at 60 V for $n$ -doped and intrinsic devices, respectively.

A manufacturing study of a 0.372 THz 8-fold (8F) helically corrugated interaction region (HCIR) and measurement of its wave dispersion characteristics are reported. This demonstrates the structure’s suitability as the electron beam / electromagnetic wave interaction region in a high power frequency tunable Gyroresonant THz source. The 8F HCIR has an eigenmode, which is ideal for broadband tuning, created by the coupling of the TE61 and TE23 modes. Maximum power handling of 14 times larger than a 3-fold HCIR at the same frequency is calculated. A TE11 to TE61 7-fold (7F) helically corrugated waveguide (HCW) mode converter, needed to measure the wave dispersion of the 8F HCIR, was designed and constructed. Negative aluminium mandrels of the 7F HCW mode converter and 8F HCIR were manufactured. Copper structures were constructed through electroforming. The measured wave dispersion of the 0.372 THz 8F HCIR agreed well with simulated dispersion characteristics.

In this article, a transmission line system for the propagation of millimeter-wave radiation is presented. The full system includes a TE11-to-TE01 mode converter, waveguide tapers, miter bends, and many straight sections. The design of each of these components is described, and the optimized simulation results are given. The mode converter shows a greater than 96% mode conversion efficiency that can be achieved over a 2% bandwidth at the ${W}$ -band. The miter bends demonstrate a transmission loss of 0.04 dB each over the same bandwidth when they are configured to introduce a mixture of higher order waveguide modes before the reflecting surface. An example transmission line system with a propagation length of 20 m, inclusive of four 90° bends with an oxygen-free high conductivity (OFHC) copper waveguide material, was studied over a 90–96-GHz frequency range and showed a 0.84-dB transmission loss at 93 GHz.

Aluminum-gallium-nitride photodetectors were successfully fabricated with micrometer-scale metal–semiconductor–metal structures and tested with ultrafast, UV laser pulses. The measurements were done with single-shot oscilloscopes. Pulse-broadening effects caused by the measurement system were systematically evaluated and reduced to resolve the intrinsic response time of the detector. The best-performing devices showed a response time of below 25 ps and dark currents below 20 pA. The devices showed linear response with the bias voltage and the laser energy.

The influence of aluminum composition (Al%) nonuniformity on AlGaN metal–semiconductor–metal photodetectors was thoroughly studied on a device level. The Al% fluctuation was precisely measured by X-ray photoelectron spectroscopy and mapped by energy dispersive spectrometry; the dislocation density was investigated by X-ray diffraction rocking curve. Both theoretical simulation and experimental testing showed a significant difference in the responsivity, dark current, and decay time in the device with different Al%. The Al% fluctuation is also a likely cause of the long decay time of the device.

A mode-matching method was developed to design a pillbox-type millimeter-wave window for a W-band gyrotron traveling-wave amplifier. With additional impedance matching sections, the pillbox window was able to achieve −40-dB reflection over a 10% frequency bandwidth. The manufactured pillbox window was achieved a maximum reflection of −15 dB. The reasons for the difference between the simulated and measured reflections were investigated.

The frequency agility of a ${W}$ -band gyrotron travelling wave amplifier (gyro-TWA) was demonstrated by the amplification of a frequency-swept input signal (chirp signal). The gyro-TWA was developed to provide high output power and wide bandwidth by using a helically corrugated interaction region. In the experiment a ~0.5-W signal of linearly swept frequency bandwidth of 1 GHz was amplified and a gain of over 30 dB was measured when driven by a 40 keV, 1.5 A, and axis-encircling electron beam.

This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) U.K. under Research Grant EP/K029746/1, and Science and Technology Facilities Council (STFC) U.K. under Research Grants ST/K006673/1 & ST/K006703/1, ST/N002326/1 & ST/N002318/1. An output coupler for a W-band high power wideband gyro-amplifier has been designed, manufactured and experimentally measured. It consists of a high performance sin2-parallel corrugated horn integrated with a broadband multi-layer window. The major design requirements are that the horn/window combination must have an input return loss lower than -30 dB over a 10 GHz bandwidth, provide a high quality output beam pattern, and operate under ultra-high vacuum conditions. The coupler converts a circular wave guide TE11 mode into the free space Laguerre Gaussian LG00 mode over the frequency band of 90–100 GHz with a measured return loss of between -30 and -40 dB and a simulated Gaussian coupling efficiency of over 99% at 94 GHz. Postprint

We show numerically that the spectrum of an optical pulse splits into multiple, widely separated, spectral bands when it arrives at a temporal boundary across which the refractive index suddenly changes. At the same time, the pulse breaks into several temporally separated pulses traveling at different speeds. The number of such pulses depends on the dispersive properties of the medium. We study the effect of second- and third-order dispersion in detail but also briefly consider the impact of other higher order terms. A temporal waveguide formed with two temporal boundaries can reflect the temporally separated pulses again and again, increasing the number of pulses trapped within the temporal waveguide.

The design of a higher order mode (HOM) input coupler for a low-THz gyrotron travelling wave amplifier (gyro-TWA) is presented. A two-branch waveguide coupler based on the even distribution of incident power is designed to couple the rectangular TE10 mode to the circular TE61 mode. The optimised tapered waveguide input coupler achieved an operating frequency range of 359 – 385 GHz, equating to a bandwidth of 7%. A prototype waveguide coupler scaled to W-band (75GHz to 110GHz) was manufactured. Vector Network Analyser (VNA) measured return loss of the HOM coupler is shown to be < 10dB at 90 – 96 GHz. A transmission and phase measurement is presented to analyse TE61 mode conversion in the coupler.

The input coupler is an important component for a microwave amplifier. In this paper, a sidewall single-hole input coupler for a W-band gyrotron traveling-wave amplifier that operates at the frequency range of 90–100 GHz was designed and measured. Instead of using a cutoff waveguide, a broadband Bragg-type reflector with a small spread in phase was optimized for use as part of the input coupler. The minimum radius of the reflector was two times the size of a cutoff waveguide, which reduced the possibility for some of the beam electrons being collected in this section and lost to the amplifier interaction region.

A Brewster window integrated into a corrugated waveguide was designed, manufactured and studied for a W-band gyrotron traveling wave amplifier. A corrugated waveguide intended to guide a quasi-plane wave ( ${\rm HE}_{11}$ ) through the window, was numerically optimized and verified by measurement using a Vector Network Analyzer. The Brewster window was measured to have a reflection lower than $-22 ~{\rm dB}$ for the ${\rm TE}_{11}$ mode over the frequency range of 85–101 GHz.

Summary form only given. A quasi-optical corrugated horn has been constructed for a W band gyro-TWA1,2 based on a cusp electron beam source3–5 with a helically corrugated interaction region6–8. The design is intended to provide an effective method to decouple the radiation and the beam so that a depressed collector energy recovery system may be used9–11. The horn converts a cylindrical TE 11 mode into a free-space TEM 00 mode over a frequency band of 84 to 104 GHz. This Gaussian mode can pass through the collector system unperturbed. This type of mode converter was chosen due to the performance advantages of greater bandwidth and continuous operation over this bandwidth. The design was optimized using a mode matching method that predicts a reflection of better than −35 dB and a Gaussian coupling efficiency of 97.8%. The prototype, which has been constructed and tested, demonstrates a close agreement with these figures

Helically corrugated waveguide has been used in various applications such as gyro-backward wave oscillators, gyro-traveling wave amplifier and microwave pulse compressor. A fast prediction of the dispersion characteristic of the operating eigenwave is very important when designing a helically corrugated waveguide. In this paper, multi-mode coupling wave equations were developed based on the perturbation method. This method was then used to analyze a five-fold helically corrugated waveguide used for X-band microwave compression. The calculated result from this analysis was found to be in excellent agreement with the results from numerical simulation using CST Microwave Studio and vector network analyzer measurements.

We present the results of an experimental study of dimensional ratios dependencies on the performance of a ballistic deflection transistor (BDT) operating in a quasi-ballistic regime. Experimental transconductance change based on geometry variations is studied for smaller and larger devices with channel width of 300 and 500 nm, respectively. Transconductance variation for a series of drain biases is also observed for a specific geometry and dimension. By means of Monte Carlo modeling we report the effect of different geometry parameters on the transfer characteristics of BDTs. The strength of the gate control in the InGaAs channel is analyzed.

High-dynamic-range, single-shot pulse shapes were measured by temporally stacking pulses in a passive, fiber-optic network. The 256 replicas were combined to produce optical shapes with a bandwidth of 10 GHz and a dynamic range of 1800:1. The high fidelity of this system enabled the characterization of arbitrary electrical pulses that were used to shape the optical pulse via an electro-optic modulator with a reduced dynamic range of about 60:1.

A novel thermionic cusp electron gun operating in the temperature-limited regime that produces a large-orbit electron beam through a nonadiabatic magnetic-field reversal was designed, analyzed, and optimized to give an electron-beam ideal for driving gyro-devices, particularly in the millimeter-to-submillimeter-wavelength range due to its small cross-sectional size. The annular-shaped axis-encircling electron beam had a beam current of 1.5 A at an acceleration potential of 40 kV, a tunable velocity ratio alpha (= vperp/vz) between one and three, an optimized axial velocity spread A.vx/vz of ~8%, and a relative alpha spread Deltakappa/alpha of ~10% at an alpha value of 1.65.

A broadband multi-layer microwave window operating in the W-band frequency range was numerically simulated, constructed and its microwave properties measured. The window acts to separate vacuum and atmosphere. It had to meet strict requirements for application in gyro-amplifiers of broadband, low reflection and high mechanical strength. The window was manufactured showing a ${- 20}~{\rm dB}$ reflection over 90–100 GHz (except in two small regions), which is in good agreement with the numerical simulation presented in this letter.

A special fiber-optic signal delivery system for 351-nm pulse-shape measurements has been developed in the contest of a large laser fusion facility. To deliver a sufficient signal and preserve an overall bandwidth of >33 GHz over a distance of /spl sim/15 m, a seven-fiber bundle of gradient-index fiber was developed, fabricated, and tested, and is now applied in routine operation.

Measuring optical pulse shapes beyond the dynamic range of oscilloscopes is achieved by temporal pulse stacking in a low-loss, passive, fiber-optic network. Optical pulses are averaged with their time-delayed replicas without introducing additional noise or jitter, allowing for high-contrast pulse-shape measurements of single-shot events. A dynamic-range enhancement of 3 bits is experimentally demonstrated and compared with conventional multishot averaging. This technique can be extended to yield an increase of up to 7 bits of additional dynamic range over nominal oscilloscope performance.

This work presents an ultrafast, interferometric electrooptic sampling system that uses a two-dimensional detector array to image the electric field present on a device. Spatial and temporal resolution are comparable to conventional "point" electrooptic sampling systems

A high-current-density, superconducting thin film will exclude a perpendicular magnetic field, produced either by a coil or a magnet. When the film is driven into the normal state by a fast optical pulse, the screening currents decay, allowing flux to enter. The process of flux entry can be observed by measuring the induced voltage across a coil closely coupled to the film. This is equivalent to the operation of an opening switch and can be used to generate fast current pulses. This concept has been experimentally demonstrated in low and high magnetic fields, corresponding to the reversible and irreversible regimes, using 500-nm-thick YBCO films and a 1.064-/spl mu/m, 150-ps pulsed laser. Theoretical analyses of current redistribution in the film associated with the transient flux motion are compared with experimental results. Prospects for applications of this concept are discussed. >

We measured in the time domain fast metal-semiconductor-metal ultraviolet photodetectors fabricated on GaN with finger width and pitch ranging from 0.3 to 5 /spl mu/m. A broad-band circuit was designed to couple out the short electrical pulse. The minimum temporal resolution between consecutive optical pulses was 26 ps. At low illumination levels, the bandwidth of the response was limited by the measurement system, not the device. At high illuminations, recovery of diode became slower, consistent with the space-charge screening effect caused by the photogenerated carriers.

The dynamic behavior of a GaAs photoconductive switch was studied with an electro-optic imaging system during the first 2 ns after optical illumination. The switch behavior changed as a function of the spatial distribution of the optical illumination. Symmetric and asymmetric illumination schemes were investigated experimentally with our electro-optic imaging system. The electric fields were significantly enhanced in the regions of low photo-carrier density. Approximately 1 ns after illumination the simple longitudinal variation of the electric field gave way to nonuniform transverse structure. The experimental results were modeled by treating the switch as an integral part of a transmission line consisting of discrete elements. The experimental results matched the predictions of the transmission-line model in terms of the electric-field enhancements and efficiency. >

Thin films of high-T/sub c/ superconducting material were used to construct two different types of opening switches: the photoresistive switch and the inductively coupled switch. The former show two distinct switching components in 8000-AA films consisting of a slow bolometric response and a faster nonbolometric response. The fast nonbolometric component had risetimes of approximately 4 ns and falltimes of approximately 6 ns. The slower component was identified to be thermal in origin and was in good agreement with the one-dimensional heat flow model used to simulate the thermal photoresponse. The inductively coupled switch is a contactless switch where the superconducting film screens the flux coupling between two coils of a transformer. The optically induced switching was achieved with risetimes of approximately 50 ns. with a multiturn output coil. Short (150-ps) laser pulses containing energies of up to 3 mJ are used in both cases to perform the switching. >

The electric field in GaAs photoconductive switches has been observed with an ultrafast electro-optic imaging system to develop complex spatial and temporal structure immediately after illumination. High-field domains form at the switch cathode as the photogenerated carriers recombine for bias fields above approximately 10 kV/cm. At these biases, the switch also remained conductive for a much longer time ( approximately 100 ns) than the material recombination time ( approximately 1 ns). A model which includes field-dependent mobility was developed to explain this data. Simulation of the electric field profile across the switch indicates that high-field domains which form at the switch cathode are the result of negative differential resistance. >

The dc null in a digital magnetic recording channel is equalized in a novel way which allows recovery of direct recorded NRZ waveforms with no coding constraints. The technique involves decision feedback equalization which can be implemented with simple RC filters. Feedback power is determined by the choice of RC and can be optimized independently of the fact of equalization. Measurements on a commercial helical-scan video transport indicate that the harmful effects of a dc null can be greatly reduced; specifically the error rate (excluding dropouts) on a 1023-bit 20-Mbit/sec pseudorandom pattern was reduced from 3 × 10-5to below 10-8.

The present study is a sequel to the authors' 1972 paper [1] where the output signal-to-noise ratio SNR o is obtained for nonadaptive differential pulse-code modulation (DPCM) systems with an arbitrary linear predictor operating on noisy digital channels. Important stability constraints for an arbitrary linear predictor are obtained. A complete analytical study of previous line-line-and-sample feedback systems is presented. The SNR o improvement over previoussample feedback is found to increase from approximately 2.6 dB for error-free channels to 4.1 dB for noisy channels. Optimization of the predictor for noisy channel usage is shown to greatly reduce the sensitivity of SNR o to variations in message and channel parameters, while use of the resulting predictor on error-free channels yields SNR o values which are almost as good as those obtained when the predictor is optimized for error-free channels. Reduction of the effects of digital channel errors on SNR o using various methods, including periodic or pseudorandom resetting are considered briefly. Hardcopy results from computer-simulated DPCM monochrome image transmission systems corroborate our analytical results.

The Fermilab Booster synchrotron requires precise injection and extraction field regulation (±0.001% repeatability). To achieve this tolerance with a current regulator requires that the current measuring device have a high degree of resolution, stability, and a high output level. To meet this criteria, a transductor using the second-harmonic magnetic modulator technique was chosen. This device is capable of better than ±0.001% measurements from dc to 15 Hz for currents from 50 to 1200 A. Small signal (~100 A peak-to-peak) response is 3 dB down at 10 KHz. The conversion gain was chosen as 1 V/121.25 A to permit operational amplifier compatible signal levels at the output.

Narrow lines and microbridge structures have been etched in sputtered superconducting films of Y-Ba-Cu-O by variations of two methods. The first uses standard photolithography followed by wet etching in weak acid. The second uses a maskless process involving focused pulsed YAG (yttrium-aluminium-garnet) laser together with a computer-controlled x-y stage to produce local ablation of the superconducting film. Issues relating to limits of resolution, annealing of films, and degradation of superconducting properties are critically discussed for the two approaches. >

An adaptive differential speech encoder was assessed using subjective evaluation procedures. The coder's adaptive quantizer was similar to the one used by Cohn and Melsa [1] and the predictor involved nonadaptive previous-sample feedback. The digital channel used to transmit the quantizer output levels was assumed error-free. Paired comparison tests were used to obtain scaled isopreference contours on the N-f plane, where N and f denote, respectively, the number of quantizer output levels and the sampling rate relative to the Nyquist rate. These contours were used to determine the subjective signal-to-noise ratio vs. f and N , maximum subjective signal-signal-to-noise ratios vs. bit rate, optimum values of f and N , and bit-rate savings which occur when entropy coding is used instead of natural coding of the quantizer output levels. Entropy coding yielded a bit rate approximately equal to threequarters that for natural coding and Nyquist-rate sampling minimized the bit rate in each case. Savings of from one to two bits occurred when ADPCM was compared with nonadaptive DPCM. The fact that our system was better than others for f \simeq 1.0 but worse for f \gsim 2.0 indicates the need to modify our quantizer adaptation algorithm as the sampling rate increases relative to the Nyqusit rate.

A new network decomposition-optimization algorithm is presented and evaluated via computer simulation. The network is first partitioned into subnetworks on the basis of lines fitted via linear regression to the node locations weighted by their traffic loads. Each node is then connected via a data line of appropriate capacity to a concentrator-multiplexer. The concentrators are then regarded as nodes, and the process is repeated as often as required. The resulting singly connected network consists of a hierarchy of concentrators whose number, capacity, location, and interconnection are selected as an inherent part of the design procedure to minimize network cost. The linear regression clustering (LRC) design procedure is evaluated via computer simulation by comparing the costs and performance of the resulting networks with that which results from use of an algorithm based on the generally applicable network design approach of iterative local optimizations or "search" procedures. The data supplied to the LRC and search algorithms include randomly generated node locations and traffic matrices, and specific (realistic) cost vs. capacity schedules for data lines and concentrators. Comparisons of network costs, queuing and transmission delays, network reliability and network design costs show the LRC algorithm to be markedly superior to the search algorithm. The paper includes a brief discussion of our results, and their implications.

Notwithstanding much effort, explicit analytical results for average message delays for asymmetric token rings are currently unavailable. We develop an excellent and relatively simple analytic approximation which readily demonstrates the effects of the many system parameters on delay-throughput performance. The approximation applies to any cyclically served network (including rings, polling networks, and unidirectional fiber optic broadcast networks) with independent time-slot arrivals and gated service. The approximation becomes exact for symmetric loading. An equation for the moment-generating function of the joint terminal service times is obtained via an imbedded Markov chain. Two matrix equations are then derived, one for the mean terminal service times of the N queues and another for the joint second central moments. A two-dimensional difference equation for the second central moments is then obtained. This difference equation together with Kleinrock's conservation law is used to approximate the mean delay at each queue. The peak approximation error is typically a few percent for various loading distributions including heavy or highly asymmetric loading, even if the statistics for message arrivals, message lengths, or walk times vary among the queues. Actual delay-throughput results are presented for various terminal loadings. Also included is a procedure to upperbound the error of our approximation for any message and walk time statistics, given the terminal utilizations.

Cost-effective design of networks linking many remote terminals to a central computer (CPU) involves use of low-speed data lines to link geographically close terminals to concentrators. The concentrators are connected via high-speed data lines to the CPU. A design algorithm based on clustering of terminals followed by optimization of location, capacity and number of concentrators in each cluster is developed and evaluated. Evaluation is based on network designs for sets of 20 randomly (uniformly) generated locations of up to 500 terminals, with specific (realistic) cost versus capacity schedules being used for data lines and concentrators. In comparison with the popular Add algorithm, our linear regression clustering (LRC) algorithm has the following advantages: 1) the total cost of the concentrators, low-speed terminal lines, and high-speed CPU lines is typically 8 percent less; 2) the average transmission time delay at the terminals is typically 40 percent less; 3) the cost of adding low-speed data lines to connect additional terminals to concentrators in existing networks is typically 50 percent less; 4) the computational cost of design is typically 20 times less for 100-terminal networks and 150 times less for 500-terminals networks. Implications of the results and suggestions for further work are discussed.

The properties of a magnetic tape recorder are viewed in terms of a digital magnetic recording/playback channel which exhibits fading (reduction of playback-signal level) and nonlinear behavior. A method is presented whereby channel nonlinearity may be quantified in a format useful for signal and receiver design. Measurements show this nonlinearity to be relatively small for symmetric two-level signals. Deep fades (dropouts) are the most significant source of errors in digital tape recordings. Fading is considered as multiplicative noise on an essentially linear channel, and measurements are made of the fade probability distribution and an associated additional time dispersion. While the fading process appears to occur relatively slowly (compared with the bit period), neither its probability distribution nor its associated dispersion encourage the use of a receiver which is able to adapt to the changing channel characteristics. Finally an attempt is made to ascribe the fading process to repeatable variations in head-tape separation, and a corresponding probability distribution for this separation is obtained. A helical-scan video recorder was used throughout the measurements because of its low cost and its potential as a high-density storage facility.

The performance of urea as a new nonlinear optical crystal for phase-matched second-harmonic generation and sum-frequency mixing in the UV is reported. Tunable radiation down to 229 nm was obtained. In particular, efficient doubling at room temperature of CW argon laser lines from 488 to 528 nm was achieved. The nonlinear coefficient of urea was measured and found to be 2.5 times that of ADP. Angle tuning curves for type II SHG, several type II frequency mixing schemes, temperature tuning data, and comparison to other nonlinear materials are presented. More accurate Sellmeier constants were also obtained using the measured angle tuning curve for type I SHG.

The rapid variation of an admittance shunting a transmission line is a well-known technique for switching microwave power. The application of a switching voltage to a ferroelectric material provides a convenient means for rapidly varying an admittance between significantly different states. A multistub transmission-reflection-type switch actuated by a switching voltage of 1000 volts has been studied. The operation of the switch depends upon the ability of a ferroelectric variable capacitor to change its capacitance upon application of a switching voltage. A change in capacitance represented by a ratio of two to one results in substantial change in the input admittance of the prototype network shunting the transmission line. The prototype network is a shunt stub and is spaced nominally at /spl lambda/ /4 intervals along the transmission line to form a multistub switch. Each shunt stub includes a ferroelectric variable capacitance which employs lead strontium titanate (Pb/sub 0.315/Sr/sub 0.685/TiO/sub 3/) as the ferroelectric material. Both theoretical and experimental curves of isolation and insertion loss vs. frequency are given for two- and three-stub versions of the switch. For switching voltages of the order of 1000 volts, ferroelectric switches with an isolation of 40 dB, an insertion loss less than 1.0 dB, and a bandwidth of 10 percent are feasible.

Existing mathematical relations between the power spectral density and the mean zero-crossing rate of an ergodic random process are used to derive relations between spectral measurements and mean zero-crossing rates of a speech signal and its derivatives. Of particular significance is the equation relating zero-crossing rates to the formant parameters of vowels and vowel-like sounds. Reasonably close agreement between measured zero-crossing rates and those calculated from spectral measurements were observed for virtually all phonemes in a variety of contextual environments. Measured zero-crossing rates and corresponding calculated values for speech and its first derivative are presented for vowels, unvoiced fricatives, and unvoiced stops, all in many different contextual environments. Unvoiced fricatives /s/, / \int /, and /f/ are shown to be distinguishable from each other solely on the basis of the zero-crossing rate of the derivative signal. Some vowels are shown to be differentiable from other vowels, although measurements other than zero-crossing rates of filtered, unfiltered, and differentiated speech are shown to be necessary for complete vowel separation. For unvoiced stop consonants, the zero-crossing rate of either the signal or its derivative is shown to be useful for classification, provided some information concerning the contextual environment is available.

Series representations, as well as exact expressions, are obtained for correlation functions and power density spectra of quantized Gaussian signals transmitted over discrete memoryless channels. The mean-square reconstruction error that results from linear filtering the received samples is expressed in terms of these power density spectra and is then examined.

The following classical problem is solved and the results are applied to minimize the mean-square error e of communication and data compression systems: "a stationary random input signal x(t) is prefiltered, corrupted by noise n(t) , sampled every T seconds, and finally postfiltered to yield output \hat{x}(t) . Let the desired output z(t) = a(t) \bigotimes x(t) where a(t) is any real function of t and \bigotimes denotes convolution. Determine the linear timeinvariant pre- and postfilters which jointly minimize \epsilon = E\{\frac{1}{T}\int^{T}_{0}[z(t) - \hat{x}(t)]^{2}dt\} subject to power constraint P = E\{[k(t) \bigotimes f(t) \bigotimes x(t)]^{2}\} .'' Operator E{\dot} denotes expected value, k(t) is a real function of t , and f(t) is the prefilter impulse response. Appropriate choice of T , P,k(t) , and n(t) makes the solution applicable to amplitude, angle, pulse-amplitude (PAM), pulse-code (PCM), and differential pulsecode (DPCM) modulation systems, and data compression systems. In this analysis no restrictions are placed on the input-signal spectrum, the noise spectrum, or the passbands of the filters; furthermore, the cross correlation between signal and noise is taken into consideration. Necessary and sufficient conditions are obtained for the jointly optimum pre- and postfilters. The performance obtainable using these optimum filters is compared with that obtainable using suboptimum filters. One suboptimum filtering scheme is derived which yields virtually the same performance as optimum filters and has the practical advantage that the filter transfer characteristics are independent of noise n(t) in many cases of interest. In applying the results to PCM, DPCM, and data compression systems, the filters, sampling rate, and quantizer are jointly optimized. The performance obtainable for various filtering schemes and various communication systems is compared with the optimum attainable as calculated from information theory.

The way in which messages are coded into sensory stimuli affects the speed and accuracy with which the messages are transmitted to a human. This paper describes experiments which were conducted in order to determine some of the properties of favorable coding schemes. The experiments showed that, in general, the stimuli should vary along many dimensions, and should assume one of two coordinate values along each dimension. When there are only two allowable coordinates per dimension, the probability of error is often independent of the actual coordinate values, provided the distance between the coordinates is well above threshold at low stimulus rates. The results of these experiments were used as a guide in coding letters of the alphabet into easily identifiable auditory and tactile stimuli. The reading rates for various coding schemes were then measured. In one code, each letter of the alphabet was represented by a unique auditory or tactile stimulus. After only 12 hours of practice, subjects were able to read English text chosen from several sources at a rate of 20 words per minute. Included in the paper is a description of a tactile-kinesthetic sensory display, and some suggestions for further research.

In a recent paper [1] the authors obtained the pre- and postfilters that minimize the mean-square error between the message x(t) and its received replica \hat{x}(t) under very general conditions on both x(t) and the communication system under consideration. The present work compares the subjective effects of systems using optimum mean-square filters (OF) with those using conventional ideal low-pass filters (LF). The systems considered are pulse-amplitude (PAM), pulse-code (PCM), and differential pulsecode (DPCM) modulation systems for voice communication. For reasons presented in this paper, the OF and LF have double-sided bandwidth equal to the sampling frequency. In PAM systems, a white Gaussian channel model is assumed, while negligible channel transmission errors are assumed for PCM and DPCM systems. The subjective evaluation, based on preference tests, shows that subjective differences between PAM systems with OF and systems with LF are insignificant and are independent of sampling frequency and channel signal-to-noise ratio. Also, for those values of sampling frequency and number of bits of quantization considered, no significant differences exist between DPCM systems with OF and systems with LF. For PCM systems, however, significant improvement in subjective performance can be achieved when OFs are used instead of LFs.

Isopreference tests were used to study the subjective effects of channel transmission errors on computer-simulated PCM and previous-sample feedback DPCM voice systems. Numerically scaled isopreference contours were obtained and plotted on planes having as axes the number of quantization bits N and the bit error probability p of the binary symmetric transmission channel. For any N and p considered, the quality of DPCM speech was found to exceed that of PCM speech. With N fixed and p decreasing, the improvement of DPCM relative to PCM increased to the point where quantization noise limited the performance of both systems. Two nonredundant codes, natural binary and folded binary, were used to encode the quantizer output. The two codes yielded comparable speech quality. The maximum subjective quality obtainable for white Gaussian and Rayleigh fading channels was determined as a function of channel capacity. Use of optimum channel encoding was shown to reduce considerably the channel capacity required to obtain a given speech quality. The subjective ratings were compared with measured system signal-to-noise ratios. Finally, the effect of the overload level of the quantizer on the system signal-to-noise ratio was determined.

A closed-form analytical expression is obtained for the output signal-to-noise ratio (SNR 0 ) of differential PCM systems operating on noisy digital channels. A procedure is then proposed for maximizing SNR 0 by joint optimization of the quantizer, predictor, and sampling rate. Several examples are considered, including one- and two-sample feedback systems excited by speech and video (Gaussian) signals. The predictor used in the DPCM system is linear and time invariant, but otherwise arbitrary. As with PCM, contributions to output signal distortion can be expressed as a sum of three separate terms resulting, respectively, from quantization errors, channel transmission errors, and mutual errors arising from interaction between quantization and channel errors. When a conventional optimum linear predictor is used, transmission errors are shown to be no more serious for DPCM than for PCM. The results show that SNR 0 for a well-designed DPCM system is considerably higher than for a well-designed PCM system operating on the same digital channel, even if the channel is noisy. In considering examples, particular attention is devoted to determining maximum values for SNR 0 and to examining the dependence of SNR 0 on predictor coefficient values, number of quantization levels, sampling rate, message statistics, and channel error probability. Implications of this dependence on system design are noted.

An RDS network contains positive resistors, ideal diodes, independent voltage sources, and independent current sources. This paper presents conditions which are necessary and sufficient for a driving-point characteristic and a transfer characteristic to be realizable by an RDS two-port terminated in a twoterminal RDS load. The sufficient conditions are established by showing how the desired two-port is constructed. The resulting network is a physically symmetric lattice. Additional necessary and sufficient conditions are found for the two-port to be grounded, and a procedure for realizing characteristics which satisfy these additional conditions is presented. Other network configurations which realize RDS driving-point and transfer functions are then considered. A simple method for determining the number of elements in the various configurations is developed. This number depends on the number and location of the breakpoints in the driving-point, transfer, and load characteristics, and can be determined without synthesizing the network.

The Fermilab booster accelerator was originally conceived for acceleration of protons with an injection energy of 200 MeV to an extraction energy of ten GeV (to the 500 GeV main accelerator). Early booster operation has been limited to eight GeV. The booster beam will be more acceptable to the main accelerator if extraction is at ten GeV, thus we are now modifying the booster magnet system for ten GeV acceleration. Regulation of the booster magnetic field for injection at 200 MeV was a task even when operating to eight GeV. This paper outlines the approach which was adopted and how it relates to the ten GeV attempt. Further, it highlights the problems encountered in the design of any AC magnet system.