Showing papers by "Rensselaer Polytechnic Institute published in 1998"

Semi-Supervised Support Vector Machines

Abstract: We introduce a semi-supervised support vector machine (S3VM) method. Given a training set of labeled data and a working set of unlabeled data, S3VM constructs a support vector machine using both the training and working sets. We use S3VM to solve the transduction problem using overall risk minimization (ORM) posed by Vapnik. The transduction problem is to estimate the value of a classification function at the given points in the working set. This contrasts with the standard inductive learning problem of estimating the classification function at all possible values and then using the fixed function to deduce the classes of the working set data. We propose a general S3VM model that minimizes both the misclassification error and the function capacity based on all the available data. We show how the S3VM model for 1-norm linear support vector machines can be converted to a mixed-integer program and then solved exactly using integer programming. Results of S3VM and the standard 1-norm support vector machine approach are compared on ten data sets. Our computational results support the statistical learning theory results showing that incorporating working data improves generalization when insufficient training information is available. In every case, S3VM either improved or showed no significant difference in generalization compared to the traditional approach.

Semiconducting Ge clathrates: Promising candidates for thermoelectric applications

Abstract: Transport properties of polycrystalline Ge clathrates with general composition Sr8Ga16Ge30 are reported in the temperature range 5 K⩽T⩽300 K. These compounds exhibit N-type semiconducting behavior with relatively high Seebeck coefficients and electrical conductivity, and room temperature carrier concentrations in the range of 1017–1018 cm−3. The thermal conductivity is more than an order of magnitude smaller than that of crystalline germanium and has a glasslike temperature dependence. The resulting thermoelectric figure of merit, ZT, at room temperature for the present samples is 14 that of Bi2Te3 alloys currently used in devices for thermoelectric cooling. Extrapolating our measurements to above room temperature, we estimate that ZT>1 at T>700 K, thus exceeding that of most known materials.

Late Cenozoic exhumation of the Cascadia accretionary wedge in the Olympic Mountains, northwest Washington State

Abstract: The apatite fission-track method is used to determine the exhumation history of the Olympic subduction complex, an uplifted part of the modern Cascadia accretionary wedge. Fission-track ages are reported for 35 sandstones from the Olympic subduction complex, and 7 sandstones and 1 diabase from the Coast Range terrane, which structurally overlies the Olympic subduction complex. Most sandstone samples give discordant results, which means that the variance in grains ages is much g reater than would be expected for radioactive decay alone. Discordance in an unreset sample is caused by a mix of detrital ages, and in a reset sample is caused by a mix of annealing properties among the detrital apatites and perhaps by U loss from some apatites. Discordant grainage distributions can be successfully interpreted by using the minimum age, which is the pooled age of the youngest group of concordant fission-track grain ages in a dated sample. The inference is that this fraction of apatites has the lowest thermal stability, and will be the first to reset on heating and the last to close on cooling. Comparison of the minimum age with depositional age provides a simple distinction between reset samples (minimum age younger than deposition) and unreset samples (minimum age older than deposition). The success of the minimum-age approach is demonstrated by its ability to resolve a well-defined age-elevation trend for reset samples from the Olympic subduction complex. Microprobe data suggest that the apatites that make up the minimum-age fraction are mostly fluorapatite, which has the lowest thermal stability for fission tracks among the common apatites. Reset minimum ages are all younger than 1 5Ma, and show a concentric age pattern; the youngest ages are centered on the central massif of the Olympic Mountains and progressively older ages in the surrounding lowlands. Unreset localities are generally found in coastal areas, indicating relatively little exhumation there. Using a stratigraphically coordinated suite of apatite fission-track ages, we estimate that prior to the start of exhumation, the base of the fluorapatite partial annealing zone was located at ~10 0°C and ~4.7 km depth. The temperature gradient at that time was 19.6 ± 4. 4°C/km, similar to the modern gradient in adjacent parts of the Cascadia forearc high. Apatite and previously published zircon fission-track data are used to determine the exhumation history of the central massif. Sedimentary rocks exposed there were initially accreted during late Oligocene and early Miocene time at depths of 12.1‐14.5 km and temperatures of ~242‐28 9°C. Exhumation began at ca. 18 Ma .A rock currently at the local mean elevation of the central massif (1204 m) would have moved through the α-damaged zircon closure temperature at about 13.7 Ma and ~10.0 km depth, and through the fluorapatite closure temperature at about 6.7 Ma and ~4. 4km depth. On the basis of age-elevation trends and paired cooling ages, we find that the exhumation rate in the central massif has remained fairly constant, ~0.75 km/m.y., since at least 14 Ma. Apatite fission-track data a re used to construct a contour map of long-term exhumation rates for the Olympic Peninsula. The average rate for the entire peninsula is ~0.28 km/m.y., which is comparable with modern erosion rates (0.18 to 0.32 km/m.y.) estimated from sediment yield data for two major rivers of the Olympic Mountains. We show that exhumation of this part of the Cascadia forearc high has been dominated by erosion and not by extensional faulting. Topography and erosion appear to have been sustained by continued accretion and thickening within the underlying Cascadia accretionary wedge. The rivers that drain the modern Olympic Mountains indicate that most of the eroded sediment is transported into the Pacific Ocean, where it is recycled back into the accretionary wedge, either by tectonic accretion or by sedimentary accumulation in shelf and slope basins. The influx of accreted sediments is shown to be similar to the outflux of eroded sediment, indicating that the Olympic segment of the Cascadia margin is currently close to a topographic steady state. The record provided by our fissiontrack data, of a steady exhumation rate for the central massif area since 14 Ma, suggests that this topographic steady state developed within several million years after initial emergence of the forearc high.

The Influence of Unity and Prototypicality on Aesthetic Responses to New Product Designs

Abstract: Unity and prototypicality are important visual aspects of product design. These design principles were operationalized by modifying line drawings of existing products. The results of four experiments provide evidence that these two factors positively affect aesthetic response. These effects were strongest when visual properties were the sole basis of judgment and when design variations were easily compared. However, they persisted when aesthetic aspects were combined with other product information and when comparing design features was difficult. The effect of unity was found to be superadditive, suggesting that it has a relational, “all-or-none” character. Finally, regression analyses show that direct effects of the design modifications on aesthetic response exist in addition to possible indirect effects that are mediated by perceived typicality.

A Composite from Poly(m‐phenylenevinylene‐co‐2,5‐dioctoxy‐p‐phenylenevinylene) and Carbon Nanotubes: A Novel Material for Molecular Optoelectronics

Abstract: As research progresses towards smaller and more efficient devices, the need to develop alternative molecular scale electronic materials becomes apparent. Integrated electronic component fabrication from organics has been recognized theoretically as the ultimate goal. In order to gain a comprehensive insight into these materials, extensive research has been carried out on conjugated carbon systems over the last few decades to optimize their optical and electrical properties. For example, doping polyacetylene with I2 has been shown to result in a large increase in conductivity compared to the pristine material. However, doping polymers tends to retard their optical properties as regards luminescence by reducing their bandgaps and introducing trapping sites such as solitons, polarons, or bipolarons. The simple lesson over the years is that if materials are to be considered for luminescence, doping should not be carried out despite the desire to improve charge transport properties. We report here the first physical adopingo, to use the traditional term, using small concentrations of multiwalled nanotubes in a conjugated luminescent polymer, poly(m-phenylenevinylene-co-2,5-dioctoxyp-phenylenevinylene) (PmPV), in a polymer/nanotube composite. This can increase electrical conductivity of the polymer by up to eight orders of magnitude. The nanotubes appear to act as nanometric heat sinks, preventing the buildup of large thermal effects, caused either optically (photobleaching) or electrically, which degrade these conjugated systems. We also report that electroluminescence was achieved from an organic light-emitting diode (LED) using the composite as the emissive layer in the device. Since initial work on conjugated systems, attempts have been made to find an area where polymers and/or fullerenes could be used as active semiconductor components. Although many new and interesting materials have been synthesized to this end, very few have found a practical application. One exception is polyphenylenevinylene (PPV), first reported by Burroughes et al. as being the light-emitting semiconductor in a Schottky diode. This encouraged scientists to study a wide variety of conjugated systems, including derivatives of this polymer, in order to optimize the efficiency of light emission from such devices. Polymers for use in LEDs must possess a number of important qualities. A high quantum yield of photoluminescence is necessary and the material must remain undoped, as dopants act as trapping sites, quenching the radiative decay of excitons. It is essential therefore to find a polymer that is reasonably conductive while maintaining its luminescent properties. Most undoped polymers possess a very low conductivity and so require high aturn-ono fields to generate sufficient carriers in order to produce the excitons, which decay radiatively. This is, in practical terms, very inefficient as fields generally induce large thermal effects, consequently causing device breakdown. There are other problems that must be addressed, but elimination of these very basic ones should substantially improve efficiencies and soon lead to applications for these polymers. The polymer used in our studies is PmPV, whose structure is a variation of the more common PPV. In this case the substitution pattern leads to dihedral angles in the chain and, according to molecular mechanics energy minimization calculations, the polymer chain tends to coil, forming a helical structure. The calculated diameter of this helix in vacuum is ca. 20 Š, whilst the pitch is ca. 6 Š. Multiwalled nanotubes were produced by the arc discharge method, resulting in multiwalled nanotubes of 20 nm average diameter and lengths between 500 nm and 1.5 mm. The nanotube powder and PPV were mixed together in toluene and sonicated briefly. It is probable that the coiled polymer conformation allows it to surround layers of nanotubes, permitting sufficiently close intermolecular proximity for p±p interaction to occur. The color change was dramatic in that the polymer has a bright yellow color while the composite, at high nanotube concentrations, possesses a deep green color. Photoluminescence studies were carried out using an Ar laser at the pump wavelength of 457 nm. Electrical conductivity was measured using a twopoint probe sandwich geometry and Pt electrodes. The LED was fabricated by casting the composite onto indium tin oxide (ITO) then sputtering an aluminum electrode on top. As the polymer structure possesses helicity, it is not surprising that it is able to wrap itself around the nanotubes and keep them suspended in solution indefinitely. The actual texture of the composite can be observed in Figure 1,

Effect of partial void filling on the lattice thermal conductivity of skutterudites

Abstract: Polycrystalline samples of antimonides with the skutterudite crystal structure with La partially filling the voids have been prepared in an effort to quantify the impact of partial void filling on the lattice thermal conductivity of these compounds. It is observed that a relatively small concentration of La in the voids results in a relatively large decrease in the lattice thermal conductivity. In addition, the largest decrease in the lattice thermal conductivity, compared to ‘‘unfilled’’ CoSb 3 is not observed near 100% filling of the voids with La, as was previously believed. This suggests a point-defect-type phonon scattering effect due to the partial, random distribution of La in the voids as well as the ‘‘rattling’’ effect of the La ions, resulting in the scattering of a larger spectrum of phonons than in the case of 100% filling. An additional benefit of partial filling in thermoelectric materials is that it may be one way of adjusting the electronic properties of these compounds. Seebeck, resistivity, Hall effect and structural data for these skutterudite compounds are also presented. @S0163-1829~98!02926-9#

Iron aluminides: present status and future prospects

Abstract: This paper constitutes a broad survey of the physical, mechanical and corrosion properties of Fe3Al alloys, as well as a review of principal processing methods. This class of alloys, once thought to be inherently brittle, is shown to possess moderate ductility, provided that mechanical testing is carried out in an inert environment. Methods to improve mechanical properties by alloying and microstructural control are described. The influence of alloying elements on corrosion and stress corrosion resistance and weldability also is reviewed.

New Nitrogen-Fixing Microorganisms Detected in Oligotrophic Oceans by Amplification of Nitrogenase (nifH) Genes.

Abstract: Oligotrophic oceanic waters of the central ocean gyres typically have extremely low dissolved fixed inorganic nitrogen concentrations, but few nitrogen-fixing microorganisms from the oceanic environment have been cultivated. Nitrogenase gene (nifH) sequences amplified directly from oceanic waters showed that the open ocean contains more diverse diazotrophic microbial populations and more diverse habitats for nitrogen fixers than previously observed by classical microbiological techniques. Nitrogenase genes derived from unicellular and filamentous cyanobacteria, as well as from the α and γ subdivisions of the class Proteobacteria, were found in both the Atlantic and Pacific oceans. nifH sequences that cluster phylogenetically with sequences from sulfate reducers or clostridia were found associated with planktonic crustaceans. Nitrogenase sequence types obtained from invertebrates represented phylotypes distinct from the phylotypes detected in the picoplankton size fraction. The results indicate that there are in the oceanic environment several distinct potentially nitrogen-fixing microbial assemblages that include representatives of diverse phylotypes.

Patterned Protein Layers on Solid Substrates by Thin Stamp Microcontact Printing

Abstract: Microcontact printing (μCP) is a new method of molecularly patterning surfaces on a micrometer scale. In this paper, we present the extension of microcontact printing to producing patterned layers of proteins on solid substrates. μCP avoids the use of strong acids and bases necessary in photolithographic patterning, allowing its use for patterning of proteins and other biological layers. We also describe the methods of thin stamp microcontact printing that allow printing of isolated features previously unattainable by microcontact printing. A solution of polylysine in borate-buffered saline was printed onto a glass coverslip, yielding micrometer scale features over an area of 4 cm2.

Surface Reconstructions and Dimensional Changes in Single-Walled Carbon Nanotubes

Abstract: Dimensional stability is crucial to possible applications of single-walled nanotubes, as their properties are linked to size and topology. We observe nanotubes responding to uniform atom loss, through surface reconstruction and drastic dimensional changes. Experiments using electron irradiation evidence nanotube diameters shrinking from similar to 1.4 to 0.4 nm. Molecular dynamics simulations show that surface reconstruction and size reduction occur through dangling bond saturation, forming nonhexagonal rings and 5-7 defects in the lattice. Nonuniform atom removal results in inhomogeneous tube deformations and local necking, and formation of linear atomic carbon chains in the nanotube body.

Self-heating in high-power AlGaN-GaN HFETs

Abstract: We compare self-heating effects in AlGaN-GaN heterostructure field effect transistors (HFETs) grown on sapphire and SiC substrates. Heat dissipation strongly affects the device characteristics soon after the application of the source-drain voltage (in less than 10/sup -7/ s). Our results show that in HFET's with the total epilayer thickness less than 1.5 /spl mu/m, the thermal impedance, /spl Theta/ is primarily determined by the substrate material and not by the material of the active layer. For our devices grown on 6H-SiC substrates, we measured /spl Theta/ of approximately 2/spl deg/C/spl middot/mm/W, which was more than an order of magnitude smaller than /spl Theta/=25/spl deg/C mm/W measured for similar AlGaN/GaN HFET's grown on sapphire. Our results demonstrate that AlGaN-GaN HFET's grown on SiC substrates combine advantages of superior electron transport properties in AlGaN/GaN heterostructures with excellent thermal properties of SiC, which should make these devices suitable for high-power electronic applications.

Size Effects in Carbon Nanotubes

Abstract: The intershell spacing of multi-walled carbon nanotubes was determined by analyzing the high resolution transmission electron microscopy images of these nanotubes. For the nanotubes that were studied, the intershell spacing ${\stackrel{^}{d}}_{002}$ is found to range from 0.34 to 0.39 nm, increasing with decreasing tube diameter. A model based on the results from real space image analysis is used to explain the variation in intershell spacings obtained from reciprocal space periodicity analysis. The increase in intershell spacing with decreased nanotube diameter is attributed to the high curvature, resulting in an increased repulsive force, associated with the decreased diameter of the nanotube shells.

Design and function of novel osteoblast-adhesive peptides for chemical modification of biomaterials

Abstract: Proactive, “next generation” dental/orthopedic biomaterials must be designed rationally to elicit specific, timely, and desirable responses from surrounding cells/tissues; for example, such biomaterials should support and enhance osteoblast adhesion (a crucial function for anchorage-dependent cells). In the past, integrin-binding peptides have been immobilized on substrates to partially control osteoblast adhesion; the present study focused on the design, synthesis, and bioactivity of the novel peptide sequence Lys-Arg-Ser-Arg that selectively enhances heparan sulfate-mediated osteoblast adhesion mechanisms. Osteoblast, but not endothelial cell or fibroblast, adhesion was enhanced significantly (p < 0.05) on substrates modified with Lys-Arg-Ser-Arg peptides, indicating that these peptides may be osteoblast- or bone cell specific. Blocking osteoblast cell-membrane receptors with various concentrations of soluble Arg-Gly-Asp-Ser peptides did not inhibit subsequent cell adhesion on substrates modified with Lys-Arg-Ser-Arg peptides, providing evidence that osteoblasts interact with Arg-Gly-Asp-Ser and with Lys-Arg-Ser-Arg peptides via distinct (i.e., integrin- and proteoglycan-mediated) mechanisms, each uniquely necessary for osteoblast adhesion. The present study constitutes an example of rational design/selection of bioactive peptides, confirms that osteoblast adhesion to substrates can be controlled selectively and significantly by immobilized peptides, and elucidates criteria and strategies for the design of proactive dental/orthopedic implant biomaterials. © 1998 John Wiley & Sons, Inc. J Biomed Mater Res, 40, 371–377, 1998.

A support vector machine approach to decision trees

Abstract: Key ideas from statistical learning theory and support vector machines are generalized to decision trees. A support vector machine is used for each decision in the tree. The "optimal" decision tree is characterized, and both a primal and dual space formulation for constructing the tree are proposed. The result is a method for generating logically simple decision trees with multivariate linear, nonlinear or linear decisions. By varying the kernel function used, the decisions may consist of linear threshold units, polynomials, sigmoidal neural networks, or radial basis function networks. The preliminary results indicate that the method produces simple trees that generalize well with respect to other decision tree algorithms and single support vector machines.

GaN based transistors for high power applications

Abstract: Unique properties of GaN and related semiconductors make them superior for high-power applications. The maximum density of the two-dimensional electron gas at the GaN/AlGaN heterointerface or in GaN/AlGaN quantum well structures can exceed 2×1013 cm−2, which is more than an order of magnitude higher than for traditional GaAs/AlGaAs heterostructures. The mobility-sheet carrier concentration product for these two dimensional systems might also exceed that for GaAs/AlGaAs heterostructures and can be further enhanced by doping the conducting channels and by using “piezoelectric” doping. Current densities over 20 A mm−1 can be reached in GaN-based high electron mobility transistors (HEMTs). These high current values can be combined with very high breakdown voltages in high-power HEMTs, which are expected to reach several thousand volts. Recent Monte Carlo simulations point to strong ballistic and overshoot effects in GaN and related materials, which should be even more pronounced than in GaAs-based compounds but at much higher electric fields. This should allow us to achieve faster switching, minimizing the power dissipation during switching events. Self-heating, which is unavoidable in power devices, raises operating temperatures of power devices well above the ambient temperature. For GaN-based devices, the use of SiC substrates allows to combine the best features of both GaN and SiC technologies; and GaN/SiC-based semiconductors and heterostructures should find numerous applications in power electronics.

Embedded and efficient low-complexity hierarchical image coder

Abstract: We propose an embedded hierarchical image coding algorithm of low complexity. It exploits two fundamental characteristics of an image transform -- the well defined hierarchical structure, and energy clustering in frequency and in space. The image coding algorithm developed here, apart from being embedded and of low complexity, is very efficient and is comparable to the best known low-complexity image coding schemes available today.© (1998) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Toward the design of intelligent traveler information systems

Abstract: The emergence of Intelligent Transportation Systems (ITS) has fostered the development of advanced traveler information systems (ATIS). These systems are designed to assist travelers in making pre-trip and enroute travel choice decisions. It is contended that while many traveler information systems are innovative and make use of cutting edge technologies, they lack real machine intelligence and therefore may be limited in their ability to service the traveling public over the long-run. The purpose of this paper is to present a vision of the next generation traveler information system, termed Intelligent Traveler Information Systems (ITIS) in which artificial intelligence techniques are drawn upon to create systems capable of providing travelers with more personalized planning assistance.

Efficient enumeration of frequent sequences

Abstract: In this paper we present SPADE, a new algorithm for fast discovery of Sequential Patterns. The existing solutions to this problem make repeated database scans, and use complex hash structures which have poor locality. SPADE utilizes combinatorial properties to decompose the original problem into smaller sub-problems, that can be independently solved in main-memory using efficient lattice search techniques, and using simple join operations. All sequences are discovered in only three database scans. Experiments show that SPADE outperforms the best previous algorithm by a factor of two, and by an order of magnitude with some pre-processed data. It also has linear scalability with respect to the number of customers, and a number of other database parameters.

Emergent fundamental pedestrian flows from cellular automata microsimulation

Abstract: In recent years cellular automata (CA) have been successfully applied to modeling traffic flow. Use of a CA for modeling pedestrian flows is examined here. A particle hopping model for a single-directional pedestrian flow over a multilane walkway is presented. This model offers the advantage of effectively capturing the behaviors of pedestrians at the micro-level while attaining realistic macro-level activity. The emergent group behavior is an outgrowth of the interaction of the rule set in simulation. The results indicate that a heuristically derived minimal rule set produces flow patterns that closely resemble the accepted fundamental diagrams. Important parameters for determining the shape of the fundamental diagrams are examined. Key rules used in a vehicular traffic CA are tested for their applicability to the pedestrian CA model.

Role of oblique convergence in the active deformation of the Himalayas and southern Tibet plateau

Abstract: Noting similarities with subduction along curved oceanic trenches and using a simple block model, we show that radial vergence evident in earthquake slip vectors along the Himalayan deformation front, east-west extension on north-trending normal faults in the Himalayas and southern Tibet, and right-lateral strike slip on the Karakorum-Jiali fault zone can all result from basal shear caused by the Indian plate sliding obliquely beneath Tibet along a gently dipping, arcuate plate boundary. Within the framework of this mechanism, the normal faults in the Himalayas and southern Tibet are not proxies for the uplift history of Tibet. The distribution and style of the faults in the Himalayas and southern Tibet suggest that the basal drag from the underthrusting Indian lithosphere extends northward beneath most of southern Tibet.

Workplace Aggression-The Iceberg beneath the Tip of Workplace Violence: Evidence of Its Forms, Frequency, and Targets

Abstract: This study investigates three hypotheses with respect to workplace aggression--behaviors performed by individuals in order to harm others with whom they work or previously worked: (1) contrary to many media reports, most aggression occurring at work setting does not involve direct, physical assaults; rather, it more typically encompasses relatively subtle (ie, covert) forms of harm-doing behavior; (2) recent changes in many organizations (eg, downsizing, increased workforce diversity) have generated_conditions that tend to increase the incidence of workplace aggression; (3) workplace aggression is perceived as occurring primarily in a downward direction within organizations so that individuals report being the victim of aggression from supervisors more often than they report aggressing against such targets while the opposite is true with respect to subordinates A survey of 452 employed persons provided relatively clear support for hypotheses 1 and 2 However, contrary to hypothesis 3, participants reported that they were more frequently the victim than the perpetrator of workplace aggression with respect to subordinates as well as supervisors INTRODUCTION Corpus Christi, Texas--A former employee opened fire Monday at a refinery inspection company, killing the owner, his wife, and three workers before fatally shooting himself "When we arrived we learned that a disgruntled employee walked in and started shooting " said Assistant Police Chief Ken Bung (Associated Press, July 12, 1995)1 Evandale, Ohio--A fired truck driver who was going after "the one that screwed me over" walked into his former employer's office Friday and shot three people to death, then calmly waited to be arrested, police said Newark, New Jersey--An ex-postal employee has confessed to slaughtering two of his former colleagues and two customers inside a Montclair, NJ post office during a robbery for rent money, officials charged "He ordered them to lie down and he just shot them," said New Jersey US Attorney Ruth Hochberg (Newsletter, March 23, 1995) Incidence such as these have recently been the focus of a tremendous amount of media attention and, at first glance, this attention appears to be fully justified Each week, an average of fifteen people are murdered at work in the United States alone, a total of more than 7,600 during the past ten years (National Institute for Occupational Safety and Health, 1993) During 1992, the last year for which complete data are currently available, the Bureau of Labor Statistics reported that 1,004 employees were murdered on the job--a rate more than one-third higher than the annual average during the 1980s In fact, homicide is currently the second leading cause of death in the workplace, with transportation accidents being the leading cause (Rigdon, 1994) Further, it has been suggested that workplace homicide is the fastest growing form of murder in the United States (Filipczak, 1993) Taken at face value, these statistics seem to suggest that current public concern with workplace violence is fully appropriate Closer examination of existing data on the occurrence of workplace violence, however, points to somewhat different conclusions First, a large majority of such violence occurs in connection with the robberies and related crimes (Bureau of Labor Statistics, 1994) In other words, contrary to popular belief, most workplace violence does not involve instances such as those described above, in which angry employees suddenly open fire on co-workers or supervisors Rather, it occurs when individuals are attacked by persons from outside their workplace who have entered it for criminal purposes Second, recent surveys indicate that, while many individuals report that they have experienced harassment from others at work--actions that create a hostile work environment but do not results in physical harm--much smaller proportions (approximately 7% and 3%, respectively, report that they have been threatened with physical harm or actually experienced such events (Northwestern National Life Insurance Company, 1993) …

Tailbiting MAP decoders

Abstract: We extend the MAP decoding algorithm of Bahl et al. (1974) to the case of tail-biting trellis codes. An algorithm is given that is based on finding an eigenvector, and another that avoids this. Several examples are given. The algorithm has application to turbo decoding and source-controlled channel decoding.

Approximate solutions to the time-invariant Hamilton-Jacobi-Bellman equation

Abstract: In this paper, we develop a new method to approximate the solution to the Hamilton–Jacobi–Bellman (HJB) equation which arises in optimal control when the plant is modeled by nonlinear dynamics. The approximation is comprised of two steps. First, successive approximation is used to reduce the HJB equation to a sequence of linear partial differential equations. These equations are then approximated via the Galerkin spectral method. The resulting algorithm has several important advantages over previously reported methods. Namely, the resulting control is in feedback form and its associated region of attraction is well defined. In addition, all computations are performed off-line and the control can be made arbitrarily close to optimal. Accordingly, this paper presents a new tool for designing nonlinear control systems that adhere to a prescribed integral performance criterion.

Asteroid 6 Hebe: The probable parent body of the H-type ordinary chondrites and the IIE iron meteorites

Abstract: — The S(IV)-type asteroid 6 Hebe is identified as the probable parent body of the H-type ordinary chondrites and of the IIE iron meteorites. The ordinary chondrites are the most common type of meteorites falling to Earth; but prior to the present study, no large mainbelt source bodies have been confirmed. Hebe is located adjacent to both the v6 and 3:1 resonances and has been previously suggested as a major potential source of the terrestrial meteorite flux. Hebe exhibits subtle rotational spectral variations, indicating the presence of some compositional variations across its surface. The silicate portion of the surface assemblage of Hebe is consistent (both in overall average and in its range of variation) with the silicate components in the suite of H-type chondrites. The high albedo of Hebe rules out a lunar-style space weathering process to produce the weakened absorption features and reddish spectral slope in the S-type spectrum of Hebe. Linear unmixing models show that a typical Ni-Fe metal spectrum is consistent with the component that modifies an H-chondrite spectrum to produce the S-type spectrum of Hebe. On the basis of the association between the H chondrites and the HE iron meteorites, our model suggests that large impacts onto the relatively metal-rich H-chondrite target produced melt bodies (sheets or pods) that differentiated to form thin, laterally extensive near-surface layers of Ni-Fe metal. Fragments of the upper silicate portions of these melt bodies are apparently represented by some of the igneous inclusions in H-chondrite breccias. Alternately, masses of metal could have been deposited on the surface of Hebe by the impact of a core or core fragment from a differentiated parent body of H-chondrite composition. Subsequent impacts preferentially eroded and depleted the overlying silicate and regolith components, exposing and maintaining large masses of metal at the optical surface of Hebe. In this interpretation, the nonmagmatic IIE iron meteorites are samples of the Ni-Fe metal masses on the surface of Hebe, whereas the H chondrites are samples from between and/or beneath the metal masses.