Search

22nd IEEE Signal Processing and Communications Applications Conference (SIU) -- APR 23-25, 2014 -- Karadeniz Teknik Univ, Trabzon, TURKEY WOS: 000356351400375 It is well known that resistive grid structures yield exponentially decreasing spatial impulse response yielding a spatial frequency response similar to that of a Gaussian filter, hence used as smoothing filters. In this paper it is shown that although the grid structure can be modeled with a recurrence (difference) equation this equation is not recursively computable but the output image can be obtained by solving the node equations of the resistive grid. However, where the image size is extremely large, it is evident that the use of any matrix inversion based algorithm is not feasible. In this paper it is shown that the solution of node equations of the resistive grid can be carried out using Jacobi iteration method which is also shown to be suitable for implementation on FPGA. IEEE, Karadeniz Tech Univ, Dept Comp Engn & Elect & Elect Engn

Bu çalışmada,         t f t t t g t t t t t        , , 0              formundaki gecikme argümentli Rayleigh tipi denklem ele alınmaktadır. Bu denklemin T-periyodik çözümlerinin varlığı üzerine yeni sonuçlar elde edilmektedir. Bu sonuçlar elde edilirken örtüşen derece teorisi kullanılmaktadır. In this paper, we discusse Rayleigh-type equation with a deviating argument of the form         t f t t t g t t t t t        , , 0.              New results are obtained on the existence of T-periodic solutions for this equation. In order to obtain these results, the coincidence degree is used.

Time delays are often instability sources and give rise to undesired oscillation and performance degradation, so it is necessary to consider delay in general congestion control schemes. In this paper, a new system model for wireless sensor networks (WSNs) is established in discrete time, subsystems including delay are presented, and the overall model is achieved by blending these subsystems. For controller synthesis, common quadratic Lyapunov functions and a new approach based on nonquadratic Lyapunov functions are utilized, and a controller is designed to stabilize each subsystem. The controller synthesis results are expressed in terms of linear matrix inequalities. Moreover, the performance of our proposed scheme is considered and the decay rate is guaranteed. Finally, a set of novel congestion control schemes is derived for WSNs, and the resulting closed-loop systems are globally asymptotically stable in case of queue length changes and the consequent delay changes. The simulation results are also presented to illustrate the effectiveness of our proposed method.

In this paper, our aim is to study some properties of the tangent bundle with a deformed complete lift metric.

The performance of the state estimation (SE) depends on the accuracy of the received measured data, as well as the parameter data of the power system. In this paper, a new algorithm is proposed for the simultaneous identification and correction of measurement and branch parameter errors (series and shunt admittances) in the power system SE problem. The proposed method uses Lagrange multipliers for the identification and correction of branch parameter errors without the need for a-priori specification of suspect parameter vectors. Erroneous measurements and branch parameter values can be corrected using the proposed method to estimate the measurement and branch parameter errors. Finally, IEEE 14-, 30-, and 57-bus test systems are used to show the validity and robustness of the proposed algorithm. Single, multiple, and simultaneous errors in the conventional measurements and branch parameters are considered for different case studies.

Nonlinear buckling of bonded double-walled boron nitride nanotubes (DWBNNTs) under combined electro-thermo--mechanical loadings based on the nonlocal piezoelasticity theory and Euler--Bernoulli beam (EBB) model is presented in this paper. Coupled DWBNNTs are embedded in an elastic medium that is simulated as a Pasternak foundation. Using the Lennard-Jones model, the van der Waals interaction between 2 layers of DWBNNTs is taken into account. Considering the von Karman geometric nonlinearity, Hamilton's principle, and charge equation, higher order governing equations are derived and solved by differential quadrature method (DQM). The detailed parametric study is conducted, focusing on the remarkable effects on the behavior of nonlinear buckling loads. The results indicated that the small-scale parameter, elastic medium, boundary conditions, electric potential, aspect ratio, and different vibration phases play an important role in the nonlinear buckling of smart elastically coupled systems. In addition, it is found that the trend of figures has good agreement with those of previous research. The results of this work could be used in the design and manufacture of nano/micro-electro--mechanical systems.

In this paper, we discuss some geometric properties of almost contact metric submersions involving symplectic manifolds. We show that the structures of quasi-K-cosymplectic and quasi-Kenmotsu manifolds are related to (1, 2)-symplectic structures. For horizontally submersions of contact CR-submanifolds of quasi-K-cosymplectic and quasi-Kenmotsu manifolds, we study the principal characteristics and prove that their total spaces are CR-product. Curvature properties between curvatures of quasi-K-cosymplectic and quasi-Kenmotsu manifolds and the base spaces of such submersions are also established. We finally prove that, under a certain condition, the contact CR-submanifold of a quasi Kenmotsu manifold is locally a product of a totally geodesic leaf of an integrable horizontal distribution and a curve tangent to the normal distribution.

This paper presents a comprehensive approach to the statistical characterization of impulse breakdown voltage under the effect of a DC sweep voltage. Several goodness-of-fit tests are applied to up-and-down test results obtained in the air-insulated rod-plane gap. Three distributions, normal, logistic, and Gumbel, are compared by means of the Kolmogorov--Smirnov goodness-of-fit test, and logistic distribution is also compared to 3-parameter Weibull distributions using the likelihood ratio test. Logistic distribution is found to be a possible alternative to normal and 3-parameter Weibull distributions.

Let M be a smooth manifold with Finsler metric F, and let TM° be the slit tangent bundle of M with a generalized Riemannian metric G, which is induced by F. In this paper, we extract many natural foliations of (TM°,G) and study some of their geometric properties. Next we use this approach to obtain new characterizations of Finsler manifolds with positive constant curvature.

Time difference of arrival (TDOA)-based methods usually achieve the most accurate source localization in passive systems. In these methods, one of the sensors is assigned as the reference and the TDOA of the others with respect to the reference is measured. The source location is estimated by processing these TDOAs. In this paper, a multireference maximum likelihood (ML)-Taylor algorithm is proposed in order to decrease the source localization error of the TDOA methods. In the proposed algorithm, TDOAs are measured assuming 2 different reference sensors, and then the ML objective function of this multireference TDOA algorithm is derived and solved using Taylor series linearization. The developed method is evaluated regarding a 2-dimensional space with some different but deterministic sensor locations. Monte Carlo simulation is used to evaluate the proposed solution. The simulation results show that the proposed method has better performance relative to the other traditional TDOA-based algorithms.

In this paper, we study the geometry and topology on the oriented Grassmann manifolds. In particular, we use characteristic classes and the Poincare duality to study the homology groups of Grassmann manifolds. We show that for k=2 or n \leq 8, the cohomology groups H*(G(k,n), R) are generated by the first Pontrjagin class, the Euler classes of the canonical vector bundles. In these cases, the Poincare duality: Hq(G(k,n), R) \to Hk(n-k)-q(G(k,n), R) can be expressed explicitly.

The main aim of this paper is to prove that the maximal operator \overset{\sim}{s}\astf:=\underset{n \in P} \sup\frac{|s nf|}{\log2 (n+1)} is bounded from the Hardy space H1/2 to the space L1/2, where s nf are Fejer means of bounded Vilenkin-Fourier series.

In this paper, we define the concept of almost product Riemannian submersion between almost product Riemannian manifolds. We introduce slant submersions from almost product Riemannian manifolds onto Riemannian manifolds. We give examples and investigate the geometry of foliations that arise from the definition of a Riemannian submersion. We also find necessary and sufficient conditions for a slant submersion to be totally geodesic.

In this paper, we develop an algorithm to solve completely the Diophantine equation F(x,y)=z2, where the quartic inhomogeneous polynomial F(x,y) with integer coefficients satisfies certain technical conditions. The procedure is an extension of the version of Runge's method given by Poulakis.

In this paper the multilevel modulation techniques of 16-Differential Amplitude Phase Shift Keying (16-DAPSK) with Orthogonal Frequency Division Multiplexing (OFDM) has been proposed with the optimum ring ratio in flat Rayleigh fading channel. By this way best Bit Error Rate (BER) performance is achieved in 16-DAPSK-OFDM systems. BER derivation and the proposed system simulation results show that optimum ring ratio is obviously different for flat Rayleigh fading channel than AWGN channel. It is concluded the proposed optimum ring ratio between [2.0,2.6] can be a promising choice to achieve optimum BER performance for 16DAPSK-OFDM in flat Rayleigh fading channels.

Date of Conference: 18-20 April 2012 In this paper, the problem of estimating the impulse responses of individual nodes in a network of nodes is dealt. It was shown by the previous work in literature that when the nodes can interact with each other, fusion based adaptive filtering approaches are more effective than handling nodes independently. Here we are proposing the use of entropy functional based optimization in the adaptive filtering stage. We tested the new method on networks under Gaussian and ε-contaminated Gaussian noise. The results show that the proposed method achieves significant improvements in the error rates in case of ε-contaminated noise. © 2012 IEEE.