Fast and stable bayesian image expansion using sparse edge priors.

Smoothness assumptions in traditional image expansion cause blurring of edges and other high-frequency content that can be perceptually disturbing. Previous edge-preserving approaches are either ad hoc, statistically untenable, or computationally unattractive. We propose a new edge-driven stochastic prior image model and obtain the maximum a posteriori (MAP) estimate under this model. The MAP estimate is computationally challenging since it involves the inversion of very large matrices. An efficient algorithm is presented for expansion by dyadic factors. The technique exploits diagonalization of convolutional operators under the Fourier transform, and the sparsity of our edge prior, to speed up processing. Visual and quantitative comparison of our technique with other popular methods demonstrates its potential and promise.     

Fuzzy and Deep Belief Network Based Malicious Vehicle Identification and Trust Recommendation Framework in VANETs

Wireless Personal Communications - This work aims to implement a clustering scheme to separate vehicles into a cluster that is based on various parameters, such as the total number of relay nodes,...     

Provisioning link layer proportional service differentiation in wireless networks with smart antennas

In this paper, we present a collision free MAC protocol for wireless networks with smart antennas that provides proportional service differentiation to various classes of traffic based on their respective bandwidth demand. The proposed protocol works for diverse physical parameters such as number of interfaces at each node, number of communication frequencies, and antenna beamwidth. To the best of our knowledge, this is the first work that provides link layer differentiated services for wireless networks with smart antennas and explores the influence of the physical parameters and network topology on the performance of the MAC layer. Ashish Deopura received his B.Tech degree in Electrical Engineering from the Indian Institute of Technology Delhi, India, in 2003, and he received his M.S. degree in Computer Systems Engineering from the University of Massachusetts Amherst, in 2005. He currently works as a Modeling Engineer for OPNET Technologies located in Bethesda, MD Professor Aura Ganz is the director of the Multimedia Networking Laboratory at the University of Massachusetts at Amherst. She has authored more than 170 journal and conference papers in the areas of multimedia wireless networks, ubiquitous computing, telemedicine, and security. She is a co-author of the book: “Multimedia Wireless Networks”, Prentice Hall, 2003. Some of her recent assignments include: general co-chair of the IEEE UWBNETS workshop, general co-chair of the IEEE BROADMED workshop, general co-chair of the Massachusetts 3rd Annual R&D Conference, keynote speaker at the NSF sponsored workshop in Mobile Computing, and invited speaker at Personal and Local Wireless Network Solutions conference, and Motorola’s Wireless Communications Futures Forum, Wireless Local Area Networks Conference. She has a PhD, MSc and BSc in Computer Science from the Technion in Israel. More details can be found at: dvd1.ecs.umass.edu/wireless.     

Benefits of the use of impairment constraint routing in optical networks

In transparent optical networks, the optical signal accumulates the effects of all physical impairments present along the path it traverses. The conventional selection of signal paths based on e.g. shortest path routing without considering the signal quality and its association with the physical impairments does not always provide the optimum solution in terms of network performance such as blocking and resource utilization. This paper proposes an impairment constraint based routing algorithm to achieve an optimal combination of physical and networking performance taking into account all physical linear impairments including noise, chromatic and polarization mode dispersion, crosstalk and filter concatenation effects in an integrated approach. The performance of a typical metropolitan area network is examined and the improvement achieved when using the proposed approach compared to the conventional shortest path routing is demonstrated.     

Properties of NixZn(1−x)Fe2O4 thick films at microwave frequencies

The thick film Ni_xZn_(1−x)Fe₂O₄ on alumina substrate was prepared by screen printing of the ferrite powder synthesized by chemical co-precipitation method using nitrate precursors. These Ni_xZn_(1−x)Fe₂O₄ thick films of varying x were characterized by X-ray diffraction, FTIR spectroscopy and SEM (scanning electron microscopy). The permittivity and permeability were measured by overlay technique. Voltage standing wave ratio method was also used to measure the dielectric constant. The permittivity was found to increase with Ni content varying between 13 and 18. The permeability was ∼3.01. The overlay technique provides an easy method for measurement of permittivity and permeability of ferrite thick film.     

Subcarrier allocation and bit loading algorithms for OFDMA-based wireless networks

Orthogonal Frequency Division Multiple Access (OFDMA) is an emerging multiple access technology. In this paper, we consider OFDMA in the context of fixed wireless networks. This paper addresses the problem of assigning subcarriers and bits to point-to-point wireless links in the presence of cochannel interference and Rayleigh fading. The objective is to minimize the total transmitted power over the entire network while satisfying the data rate requirement of each link. We formulate this problem as a constrained optimization problem and present centralized algorithms. The simulation results show that our approach results in an efficient assignment of subcarriers and transmitter power levels in terms of the energy required for transmitting each bit of information. However, centralized algorithms require knowledge of the entire network topology and channel characteristics of every link. In a practical scenario, that would not be the situation and there is a need for distributed rate allocation algorithms. To address this need, we also present a distributed algorithm for allocating subcarriers and bits in order to satisfy the rate requirements of the links.     

Light‐Induced Plasmon‐Assisted Phase Transformation of Carbon on Metal Nanoparticles

Highly localized light‐induced phase transformation of electron beam induced deposited carbon nanostructures (dots and squares) on noble metal surfaces is reported. The phase transformation from the amorphous phase to the disordered graphitic phase is analyzed using the characteristic Raman signatures for amorphous and graphitized carbon and conductive force microscopy. The extent of the transformation is found to be largely dependent on the plasmon absorption properties of the underlying metal film. It is observed that the amorphous carbon deposits on the silver films consisting of 12 nm particles with the plasmon absorption near the laser excitation wavelength (514 nm), undergo fast graphitization to a nanocrystalline or a disordered graphitic phase. This transformation results in the formation of a highly conductive carbon/metal interface with at least seven orders of magnitude lower electrical resistivity than the initial insulating interface. It is suggested that the fast graphitization of nanoscale carbon deposits might serve as an efficient path for the formation of complex patterned nanoscale metal‐carbon interconnects with high electrical conductivity.     

Advances in pulsed-laser-deposited AIN thin films for high-temperature capping, device passivation, and piezoelectric-based RF MEMS/NEMS resonator applications

In this paper we report recent advances in pulsed-laser-deposited AIN thin films for high-temperature capping of SiC, passivation of SiC-based devices, and fabrication of a piezoelectric MEMS/NEMS resonator on Pt-metallized SiO₂/Si. The AlN films grown using the reactive laser ablation technique were found to be highly stoichiometric, dense with an optical band gap of 6.2 eV, and with a surface smoothness of less than 1 nm. A low-temperature buffer-layer approach was used to reduce the lattice and thermal mismatch strains. The dependence of the quality of AlN thin films and its characteristics as a function of processing parameters are discussed. Due to high crystallinity, near-perfect stoichiometry, and high packing density, pulsed-laser-deposited AlN thin films show a tendency to withstand high temperatures up to 1600°C, and which enables it to be used as an anneal capping layer for SiC wafers for removing ion-implantation damage and dopant activation. The laser-deposited AlN thin films show conformal coverage on SiC-based devices and exhibit an electrical break-down strength of 1.66 MV/cm up to 350°C when used as an insulator in Ni/AlN/SiC metal-insulator-semiconductor (MIS) devices. Pulsed laser deposition (PLD) AlN films grown on Pt/SiO₂/Si (100) substrates for radio-frequency microelectrical and mechanical systems and nanoelectrical and mechanical systems (MEMS and NEMS) demonstrated resonators having high Q values ranging from 8,000 to 17,000 in the frequency range of 2.5–0.45 MHz. AlN thin films were characterized by x-ray diffraction, Rutherford backscattering spectrometry (in normal and oxygen resonance mode), atomic force microscopy, ultraviolet (UV)-visible spectroscopy, and scanning electron microscopy. Applications exploiting characteristics of high bandgap, high bond strength, excellent piezoelectric characteristics, extremely high chemical inertness, high electrical resistivity, high breakdown strength, and high thermal stability of the pulsed-laser-deposited thin films have been discussed in the context of emerging developments of SiC power devices, for high-temperature electronics, and for radio frequency (RF) MEMS.     

A Recursive Algorithm for Computing Exact Reliability Measures

An algorithm is presented to find source-to-K-terminal reliability in a directed graph with independent arc failures. The algorithm is based on a discrete-time Markov chain with two absorbing states. The Markov chain has an upper triangular transition probability matrix, thus the probability of absorption in a state can be found by back-substitution. We show: 1) The source-to-K-terminal reliability is the probability of absorption in a particular absorbing state; 2) The time until absorption can be used as an alternative reliability measure; and 3) The algorithm can be used to find a third reliability measure called the degree of connectedness.     

Ba-M+Y (Ba4−XCo2+XFe36O60) complex hexaferrite: structural,electrical, magnetic,and enhanced microwave absorbing properties over 8–18 GHz frequency

Journal of Materials Science: Materials in Electronics - Hexaferrites are a class of materials that have practical applications as microwave absorbing materials (MAMs) and radar absorbing materials...