Spl Tau Research Articles

Satisfiability, Branch-Width and Tseitin tautologies

For a CNF /spl tau/, let w/sub b/(/spl tau/) be the branch-width of its underlying hypergraph. In this paper we design an algorithm for solving SAT in time n/sup O(1)/2/sup O(w(b)(/spl tau/))/. This in particular implies a polynomial algorithm for testing satisfiability on instances with tree-width O(log n). Our algorithm is a modification of the width based automated theorem prover (WBATP) which is a popular (at least on the theoretical level) heuristic for finding resolution refutations of unsatisfiable CNFs. We show that instead of the exhaustive enumeration of all provable clauses, one can do a better search based on the Robertson-Seymour algorithm for approximating the branch-width of a graph. We call the resulting procedure Branch-Width Based Automated Theorem Prover (BWBATP). As opposed to WBATP, it always produces regular refutations. Perhaps more importantly, the running time of our algorithm is bounded in terms of a clean combinatorial characteristic that can be efficiently approximated, and that the algorithm also produces, within the same time, a satisfying assignment if /spl tau/ happens to be satisfiable. In the second part of the paper we investigate the behavior of BWBATP on the Well-studied class of Tseitin tautologies. We argue that in this case BWBATP is better than WBATP. Namely, we show that its running time on any Tseitin tautology /spl tau/ is |/spl tau/|/sup O(1)/. 2/sup O(w(/spl tau//spl boxvr/O))/ as opposed to the obvious bound n/sup O(w(/spl tau//spl boxvr/O))/ provided by WBATP. This in particular implies that Resolution is automatizable on those Tseitin tautologies for which we know the relation w(/spl tau//spl boxvr//spl phi/) /spl les/ O(log S(/spl tau/)). We identify one such subclass and prove partial results toward establishing this relation for larger classes of graphs.

Magnetostatic surface wave bright soliton propagation in ferrite-dielectric-metal structures

Magnetostatic surface wave (MSSW) bright solitons in a ferrite-dielectric-metal (FDM) structure have been studied experimentally and numerically in the framework of the nonlinear Schro/spl uml/dinger equation. Attention was focused on the influence of the parametric instability on the soliton formation and propagation. We also discussed the contribution of the nonsolitary (dispersive wave) part of the MSSW pulse on the soliton propagation, to show that their mutual interference leads to the leveling off or to the appearance of some peaks in the MSSW pulse output versus the input amplitude. We have also shown that for MSSW pulses with rectangular shape, the linear pulse compression caused by an induced phase modulation of the input pulse must be taken into account. Experiments were performed on FDM microstrip structures loaded by a 14-/spl mu/m-thick yttrium iron garnet film, separated from the ground metal by an air gap with thickness h/sub 1//spl ap/100 /spl mu/m or h/sub 2//spl ap/200 /spl mu/m. It was found experimentally for MSSW with wavelength /spl lambda//spl ap/h that the modulation instability leads to soliton formation for rectangular input pulses with duration /spl tau/ less than the characteristic transient time t/sup */ needed for the onset of the parametric instability, while pulses with /spl tau//spl ges/t/sup */ are mainly subjected to parametric instability. The measured threshold amplitudes for parametric and modulation instabilities are in agreement with the theoretical predictions. An influence of additional pumping in the form of both continuous-wave and pulsed signals on the soliton formation was studied. It was shown that an additional pumping signal with duration /spl tau//spl ges/t/sup */, and amplitude above the threshold of the parametric instability, suppressed the MSSW soliton. Numerical modeling of the pulsewidth dependence on the microwave power during the propagation in the FDM structure yields results that are in agreement with the experimental observations. Moreover, pulse narrowing due to the induced phase modulation of the input pulse was numerically predicted. All of these effects are in agreement with the experimental findings.

Low-Power Logic Circuit and SRAM Cell Applications With Silicon on Depletion Layer CMOS (SODEL CMOS) Technology

In this paper, the switching performance of Silicon on Depletion Layer CMOS (SODEL CMOS) is investigated with a view to realizing high-speed and low-power CMOS applications. Thanks to smaller parasitic capacitance, the propagation delay time (/spl tau//sub pd/) in SODEL CMOS has been improved by up to 25% compared to that of conventional bulk CMOS in five stacked nFET inverters at the same V/sub dd/. It is also confirmed that about 30% better power-delay product can be realized at the same /spl tau//sub pd/ with reduced V/sub dd/ in SODEL CMOS. In SRAM cell applications, SODEL CMOS shows high Static Noise Margin (SNM) of /spl sim/95 mV at V/sub dd/=0.6 V. Smaller bitline delay is expected and confirmed in SODEL CMOS SRAM by SPICE simulations. Latch-up immunity for /spl alpha/-particle irradiation in SODEL CMOS was also found to be comparable to that of conventional bulk CMOS. Therefore, SODEL CMOS device and circuit technology is expected to provide a better solution for low-power system-on-a-chip (SoC).

Characteristics of a Novel Kind of Miniaturized Cavity-Cell Assembly for Rubidium Frequency Standards

A novel kind of cavity-cell assembly for rubidium frequency standards has been developed. The assembly utilizes a new magnetron-type microwave cavity. The cavity is of small size, easily manufactured with high machining precision, and the cavity frequency can be easily adjusted. Either the integrated filter technique (IFT) or the separated filter technique (SFT) can be used in this cavity-cell assembly. The assembly has a high signal-to-noise ratio. The frequency stabilities of rubidium atomic frequency standards with this assembly are 4/spl times/10/sup -12///spl tau//sup 1/2/ for IFT and 2.7/spl times/10/sup -12///spl tau//sup 1/2/ for SFT, and its temperature coefficient and light shift are very low.

High-speed automatic microscopy for real time tracks reconstruction in nuclear emulsion

The Oscillation Project with Emulsion-tRacking Apparatus (OPERA) experiment will use a massive nuclear emulsion detector to search for /spl nu//sub /spl mu///spl rarr//spl nu//sub /spl tau// oscillation by identifying /spl tau/ leptons through the direct detection of their decay topology. The feasibility of experiments using a large mass emulsion detector is linked to the impressive progress under way in the development of automatic emulsion analysis. A new generation of scanning systems requires the development of fast automatic microscopes for emulsion scanning and image analysis to reconstruct tracks of elementary particles. The paper presents the European Scanning System (ESS) developed in the framework of OPERA collaboration.

Collector-pedestal InGaAs/InP DHBTs fabricated in a single-growth, triple-implant process

This letter reports InP/In/sub 0.53/Ga/sub 0.47/As/InP double heterojunction bipolar transistors (DHBTs) employing an N/sup +/ subcollector and N/sup +/ collector pedestal-formed by blanket Fe and patterned Si ion implants, intended to reduce the extrinsic collector-base capacitance C/sub cb/ associated with the device footprint. The Fe implant is used to compensate Si within the upper 130 nm of the N/sup +/ subcollector that lies underneath the base ohmic contact, as well as compensate the /spl sim/1-7/spl times/10/sup -7/ C/cm/sup 2/ surface charge at the interface between the indium phosphide (InP) substrate and the N/sup $/collector drift layer. By implanting the subcollector, C/sub cb/ associated with the base interconnect pad is eliminated, and when combined with the Fe implant and selective Si pedestal implant, further reduces C/sub cb/ by creating a thick extrinsic collector region underneath the base contact. Unlike previous InP heterojunction bipolar transistor collector pedestal processes, multiple epitaxial growths are not required. The InP DHBTs here have simultaneous 352-GHz f/sub /spl tau// and 403-GHz f/sub max/. The dc current gain /spl beta//spl ap/38, BV/sub ceo/=6.0 V, BV/sub cbo/=5.4 V, and I/sub cbo/<50 pA at V/sub cb/=0.3 V.

Experimental Demonstration of a Compact and High-Performance Laser-Pumped Rubidium Gas Cell Atomic Frequency Standard

The authors present a compact high-performance laser-pumped Rubidium atomic frequency standard exploiting the advantages of laser optical pumping for improved stability. The clock is based on an industrial Rb clock with the lamp assembly removed and optically pumped by light from a compact frequency-stabilized laser head. The modification of the buffer gas filling in the clock resonance cell reduces instabilities on medium-term timescales arising from the ac Stark effect and temperature variations. The frequency stability of the demonstrator clock was measured to be better than 4/spl times/10/sup -12//spl tau//sup -1/2/ up to 10/sup 4/ s, limited by the local oscillator (LO) quartz and RF loop electronics. Long-term drifts under atmosphere amount to 2-6/spl times/10/sup -13//day only, comparable to or lower than that for lamp-pumped clocks under similar conditions. Typical signal contrasts lie at around 20%, corresponding to a shot-noise limit for the short-term stability of 2/spl times/10/sup -13//spl tau//sup -1/2/. The results clearly demonstrate the feasibility of a laser-pumped Rb clock reaching <1/spl times/10/sup -12//spl tau//sup -1/2/ in a compact device (< 2 L, 2 kg, 20 W), given the optimization of the implemented techniques. Compact high-performance clocks of this kind are of high interest for space applications such as telecommunications, science missions, and future generations of satellite navigation systems [GPS, global orbiting navigation satellite system (GLONASS), European satellite navigation system (GALILEO)].

Improved VBIC model for SiGe HBTs with an unified model of heterojunction barrier effects

An improved bipolar transistor model considering heterojunction barrier effect (HBE) in SiGe double heterojunction bipolar transistors is developed. The effect of barrier formation due to high level injection, which is related to the rapid degradations of the dc current gain (/spl beta/) and cutoff frequency (f/sub T/), is carefully investigated and analyzed. As the collector current becomes high, the conduction band barrier is induced and increased. It causes the saturation of collector current (J/sub C/) due to the blocking of carrier transport, the sharp increase of base transit time (/spl tau//sub B/) due to the additional charge storage, the increase of base current (J/sub B/) due to the increased recombination, and the decrease of intrinsic base resistance (R/sub bi/) due to the increased charge and base pushout. Those phenomena are included into a vertical bipolar intercompany model (VBIC) compact model by employing a unified model of the HBE on J/sub C/, J/sub B/, /spl tau//sub B/, and R/sub bi/. Furthermore, portions of /spl tau//sub B/ and R/sub bi/ from the Kirk effect itself are modeled according to the high current model description and the new formulation of widened base, respectively. A full extraction of parameters has been performed and the modified VBIC model is applied. The modeling accuracy is significantly improved at the high current region for the dc and RF characteristics.

General thermal force model with experimental studies

Thermal force has proven to be a useful concept for managing the package-level thermal placement problem. However, the previous thermal force model is based on insulated edge boundary condition; thus it is only suitable to face-cooled packages. A general thermal force model is proposed to extend the applicability of the thermal force to cover general cooling conditions by introducing the concept of the heat transparency of a boundary /spl tau/. By managing /spl tau/, the present method generates a series of thermal-force-equilibrium placements fitting to situations from face-cooled to edge-cooled packages. Experimental results indicate that for generating the reliability-best placements, the best /spl tau/'s are in the range 0 /spl les/ /spl tau/ /spl les/ 0.05 for face-cooled packages and in the range 0.1 /spl les/ /spl tau/ /spl les/ 0.25 for the volumetric-cooled packages. For edge-cooled packages, the best /spl tau/'s are the largest values of /spl tau/ that can achieve convergent thermal-force-equilibrium placements.

On the Impact of Soft Handoff Threshold and Maximum Size of the Active Group on BS Transmit Power in the UMTS System

This paper investigates the effect of two control parameters of the Universal Mobile Telecommunications System (UMTS) handoff algorithm on the total peak base station (BS) transmit power required per mobile station (MS) to satisfy a prescribed outage probability. Denoting the set of BSs supporting an MS in soft handoff (SH) as the active group, the first parameter n is the maximum allowed size of the active group. The second, denoted /spl tau/, is the upper limit on the difference between the path loss of the dominant BS and that of any other member of the active group, where the term "dominant" refers to the BS with the lowest path loss to the MS. It was assumed that the MS is equipped with a Rake receiver capable of performing "maximal ratio combining" of the signals it receives from the transmitting BSs. General analytical derivations along with results derived for specific situations through numerical integration are presented.

Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information

This paper considers the model problem of reconstructing an object from incomplete frequency samples. Consider a discrete-time signal f/spl isin/C/sup N/ and a randomly chosen set of frequencies /spl Omega/. Is it possible to reconstruct f from the partial knowledge of its Fourier coefficients on the set /spl Omega/? A typical result of this paper is as follows. Suppose that f is a superposition of |T| spikes f(t)=/spl sigma//sub /spl tau//spl isin/T/f(/spl tau/)/spl delta/(t-/spl tau/) obeying |T|/spl les/C/sub M//spl middot/(log N)/sup -1/ /spl middot/ |/spl Omega/| for some constant C/sub M/>0. We do not know the locations of the spikes nor their amplitudes. Then with probability at least 1-O(N/sup -M/), f can be reconstructed exactly as the solution to the /spl lscr//sub 1/ minimization problem. In short, exact recovery may be obtained by solving a convex optimization problem. We give numerical values for C/sub M/ which depend on the desired probability of success. Our result may be interpreted as a novel kind of nonlinear sampling theorem. In effect, it says that any signal made out of |T| spikes may be recovered by convex programming from almost every set of frequencies of size O(|T|/spl middot/logN). Moreover, this is nearly optimal in the sense that any method succeeding with probability 1-O(N/sup -M/) would in general require a number of frequency samples at least proportional to |T|/spl middot/logN. The methodology extends to a variety of other situations and higher dimensions. For example, we show how one can reconstruct a piecewise constant (one- or two-dimensional) object from incomplete frequency samples - provided that the number of jumps (discontinuities) obeys the condition above - by minimizing other convex functionals such as the total variation of f.

Digital filter design using root moments for sum-of-all-pass structures from complete and partial specifications

This paper is concerned with the development of digital filter design procedures for transfer functions in the form of sum-of-all-pass in which the requirements may be partially specified. Specifically, the requirements for a digital filter or equalizer in amplitude A(/spl theta/), or phase /spl phi/(/spl theta/), or possibly group-delay response /spl tau/(/spl theta/), may be specified from measurements over a limited a set of frequencies {/spl theta//sub 1/,/spl theta//sub 2/}. The problem is to develop techniques for the design a transfer function H(z) satisfying these specifications and constrained to be in the form of a sum-of-two-all-pass functions. The proposed solution is based on the use of root moments. The companion problem concerned with the estimation of the orders of the required all-pass filters is also examined and a solution proposed based on the same context of root moments.

Analysis of Transit Times and Minority Carrier Mobility in n-p-n 4H-SiC Bipolar Junction Transistors

An analysis of the transit times and minority carrier mobility in n-p-n 4H-SiC RF bipolar junction transistors is presented. These parameters were extracted from small signal RF measurements on 4H-SiC RF transistors with three different base thicknesses: 100, 140, and 200 nm. The study shows that the room temperature minority carrier electron mobility is 215 cm/sup 2//V/spl middot/s for a base Al doping of N/sub B/=4/spl times/10/sup 18/ cm/sup -3/. The analysis reveals that the collector charging time /spl tau//sub C/ and the parasitic charging time /spl tau//sub P/ from the capacitance between metal pads and the underlying collector region have a significant effect on the transistors RF performance. The calculated RF gain is in good agreement with the measured results.

Torque multiplication and stable range tradeoff in parallel plate angular electrostatic actuators with fixed DC bias

The nonlinear torque-voltage characteristics in two-terminal electrostatic actuators can be utilized to magnify the torque generated by a drive voltage applied to one electrode if a fixed dc bias is applied to the other. The resulting torque is enhanced by torque gain factor G/sub /spl tau//>1, and the drive voltage is effectively multiplied by voltage gain factor G/sub V/>1 compared to that of an actuator with no dc bias. These gain factors are generated at the expense of a reduced stable range. In this paper, we study and determine experimentally the tradeoff between torque and voltage gains versus stable range for one-dimensional (1-D), three-terminal, parallel-plate angular electrostatic actuators under dc bias. Simple approximate analytical relations are derived for voltage and torque gain as functions of applied dc bias voltage. We demonstrate that for voltage gains of 2-4, the angular range is marginally reduced. [1361].

High-performance dual-gate carbon nanotube FETs with 40-nm gate length

We report on a high-performance back-gated carbon nanotube field-effect transistor (CNFET) with a peak transconductance of 12.5 /spl mu/S and a delay time per unit length of /spl tau//L=19 ps//spl mu/m. In order to minimize the parasitic capacitances and optimize the performance of scaled CNFETs, we have utilized a dual-gate design and have fabricated a 40-nm-gate CNFET possessing excellent subthreshold and output characteristics without exhibiting short-channel effects.

Inspection of intrinsic critical currents for spin-transfer magnetization switching

We examined the relationships between critical current, I/sub c/, and switching time, /spl tau//sub p/, for spin-transfer switching in two regions: (region I) /spl tau//sub p//spl Gt//spl tau//sub 0/, where thermal switching is accompanied and (region II) /spl tau//sub p/< several tens times /spl tau//sub 0/, where /spl tau//sub 0/ is the attempt time for thermal switching (/spl ap/1 ns). We estimated I/sub c0/, defined as the intrinsic I/sub c/ at 0 K, for both regions and confirmed experimentally that those I/sub c0/ coincided with each other at room temperature (RT). The value of I/sub c/ at /spl tau//sub p/=1 ns, measured with microwaves, was approximately 1.6 times the I/sub c0/. This suggested that we use at least two times I/sub c0/ as the writing currents of magnetic memory devices for nsec spin-transfer switching at RT. Although I/sub c0/ for both regions were defined as I/sub c/ at 0 K (I/sub c//sup 0K/) in theory, they showed temperature dependence at low temperatures; |I/sub c0/| for region I increased with decreasing temperature, and the estimated I/sub c//sup 0K/ was approximately three times I/sub c0/ for RT. This temperature dependence was quite different from that for region II.

Spin-transfer switching current distribution and reduction in magnetic tunneling junction-based structures

Spin transfer switching current distribution within a cell and switching current reduction were studied at room temperature for magnetic tunnel junction-based structures with resistance area product (RA) ranged from 10 to 30 /spl Omega/-/spl mu/m/sup 2/ and TMR of 15%-30%. These were patterned into current perpendicular to plane configured nanopillars having elliptical cross sections of area /spl sim/0.02 /spl mu/m/sup 2/. The width of the critical current distribution (sigma/average of distribution), measured using 30 ms current pulse, was found to be 3% for cells with thermal factor (KuV/k/sub B/T) of 65. An analytical expression for probability density function p(I/I/sub c0/) was derived considering a thermally activated spin transfer model, which supports the experimental observation that the thermal factor is the most significant parameter in determining the within-cell critical current distribution. Spin-transfer switching current reduction was investigated through enhancing effective spin polarization factor /spl eta//sub eff/ in magnetic tunnel junction-based dual spin filter (DSF) structures. The intrinsic switching current density (J/sub c0/) was estimated by extrapolating experimental data of critical current density (J/sub c/) versus pulse width (/spl tau/), to a pulse width of 1 ns. A reduction in intrinsic switching current density for a dual spin filter (DSF: Ta/PtMn/CoFe/Ru/CoFeB/Al2O3/CoFeB/spacer/CoFe/PtMn/Ta) was observed compared to single magnetic tunnel junctions (MTJ: Ta/PtMn/CoFe/Ru/CoFeB/Al2O3/CoFeB/Ta). J/sub c/ at /spl tau/ of 1 ns (/spl sim/J/sub c0/) for the MTJ and DSF samples were 7/spl times/10/sup 6/ and 2.2/spl times/10/sup 6/ A/cm/sup 2/, respectively, for identical free layers. Thus, a significant enhancement of the spin transfer switching efficiency is seen for DSF structure compared to the single MTJ case.

Optical molasses loaded from a low-velocity intense source of atoms: an atom source for improved atomic fountains

We demonstrate the efficient capture of cesium atoms from a low-velocity intense source (LVIS) of atoms into an optical molasses. The high load rates that are achievable with the technique can potentially improve the stability of and reduce collisional shifts in atomic fountain clocks. An LVIS is an atomic beam created from a small hole in one of the retroreflectors in an otherwise ordinary magnetooptical trap (MOT). Our typical LVIS flux was 10/sup 10/ atoms/s. The asymptotic value for the number of atoms captured in the optical molasses was 1.1(1) /spl times/ 10/sup 9/ atoms, and the fill time constant was /spl tau/ = 290(30) ms. The initial molasses fill rate was R/sub t=0/ = 3.8(5) /spl times/ 10/sup 6/ atoms/ms. At this rate, it would require 24 ms to capture and launch 10/sup 7/ state-selected atoms in an atomic fountain. The fill rate at short times indicates that approximately 40% of the LVIS atoms were being captured in the optical molasses.

Investigation of the asymmetric characteristics and temperature effects of CdZnTe detectors

The results of our newest depth sensing coplanar grid CdZnTe detectors will be discussed. These detectors, which utilize the third-generation coplanar anode design, were fabricated by Baltic Scientific Instruments, Ltd., using crystals acquired from Yinnel Tech, Inc., having dimensions of 1.5 cm /spl times/1.5cm/spl times/ 0.9 cm and 1.5 cm /spl times/ 1.5cm /spl times/ 0.95 cm. Employing various compensation techniques, an energy resolution of about 1.7% full width at half maximum (FWHM) was achieved for 662-keV gamma rays. We observed that the spectral performance was highly dependent upon the biasing orientation of the detector. The /spl mu//sub e//spl tau//sub e/ product for one of the crystals was measured and it was found to be a high value of 1.13 /spl times/ 10/sup -2/ cm/sup 2//V. A study of the spectroscopic performance of CdZnTe for reduced temperature operation was implemented and we observed degraded performance for temperatures below 0/spl deg/C and a complete loss of spectral information at -30/spl deg/C.

Transistor and circuit design for 100-200-GHz ICs

Compared to SiGe, InP HBTs offer superior electron transport properties but inferior scaling and parasitic reduction. Figures of merit for mixed-signal ICs are developed and HBT scaling laws introduced. Device and circuit results are summarized, including a simultaneous 450 GHz f/sub /spl tau// and 490 GHz f/sub max/ DHBT, 172-GHz amplifiers with 8.3-dBm output power and 4.5-dB associated power gain, and 150-GHz static frequency dividers (a digital circuit figure-of-merit for a device technology). To compete with advanced 100-nm SiGe processes, InP HBTs must be similarly scaled and high process yields are imperative. Described are several process modules in development: these include an emitter-base dielectric sidewall spacer for increased yield, a collector pedestal implant for reduced extrinsic C/sub cb/, and emitter junction regrowth for reduced base and emitter resistances.