Design of 2D FIR and IIR Digital Filters with Canonical Signed Digit Coefficients Using Singular Value Decomposition and Genetic Algorithms

In this paper a design approach for two-dimensional finite impulse and infinite impulse response digital filters with canonic signed digit (CSD) coefficients based on singular value decomposition and genetic algorithms (GAs) is presented. The proposed technique uses a new chromosome coding scheme to alleviate the problem encountered by many GAs during crossover and mutation which destroys the CSD property. Examples are given to demonstrate the usefulness of the proposed technique. A comparative study carried out with some of the existing techniques indicates the high throughput property of the proposed technique.     

Exploiting symmetries for scaling loopy belief propagation and relational training

Judging by the increasing impact of machine learning on large-scale data analysis in the last decade, one can anticipate a substantial growth in diversity of the machine learning applications for “big data” over the next decade. This exciting new opportunity, however, also raises many challenges. One of them is scaling inference within and training of graphical models. Typical ways to address this scaling issue are inference by approximate message passing, stochastic gradients, and MapReduce, among others. Often, we encounter inference and training problems with symmetries and redundancies in the graph structure. A prominent example are relational models that capture complexity. Exploiting these symmetries, however, has not been considered for scaling yet. In this paper, we show that inference and training can indeed benefit from exploiting symmetries. Specifically, we show that (loopy) belief propagation (BP) can be lifted. That is, a model is compressed by grouping nodes together that send and receive identical messages so that a modified BP running on the lifted graph yields the same marginals as BP on the original one, but often in a fraction of time. By establishing a link between lifting and radix sort, we show that lifting is MapReduce-able. Still, in many if not most situations training relational models will not benefit from this (scalable) lifting: symmetries within models easily break since variables become correlated by virtue of depending asymmetrically on evidence. An appealing idea for such situations is to train and recombine local models. This breaks long-range dependencies and allows to exploit lifting within and across the local training tasks. Moreover, it naturally paves the way for the first scalable lifted training approaches based on stochastic gradients, both in an online and a MapReduced fashion. On several datasets, the online training, for instance, converges to the same quality solution over an order of magnitude faster, simply because it starts optimizing long before having seen the entire mega-example even once.     

Robustification of input redundant feedback systems using robust actuator weighting in the control allocation problem

In this article, a new methodology for robust actuator weighting in the control allocation (CA) problem of input redundant feedback systems is addressed. The methodology is based on the control structural properties of the plant which were previously used for control configuration selection. Robust performance (RP) measures including H _∞ norm and structured singular value of the closed-loop system are used in this article. The capability of the approach is proven with application to lateral dynamics control of the vehicle over-actuated with front and rear steering systems. Employing the RP measures, it is concluded that the vehicle feedback control with front steering angles gives superior RP properties in comparison with the feedback loop of the rear steering angles. Based on these results, the penalty weightings in the cost function of the CA unit are determined. Simulation results based on nonlinear seven degrees of freedom vehicle handling model show that the selection of penalty weightings in the CA unit based on the RP properties of the control inputs (front and rear steering angles) improves the RP of the closed-loop.     

Task-space control of robots using an adaptive Taylor series uncertainty estimator

An uncertainty estimation and compensation can improve the performance of control systems due to structured and unstructured uncertainty. This paper presents a robust task-space control approach using an adaptive Taylor series uncertainty estimator for electrically driven robot manipulators. It is worth noting that not only the lumped uncertainty is estimated and employed in the indirect form of robust controller, but also the upper bound of approximation error is estimated to form a robustifying term and the asymptotic convergence of tracking error and its time derivative are proven based on stability analysis. Finally, the effectiveness of the proposed controller is shown through simulation and comparison with two valuable control schemes applied on the Selective Compliance Assembly Robot Arm (SCARA) robot manipulator.     

A high-performance switch fabric for integrated circuit and packet switching

This paper describes an architecture for a high-performance switching fabric that can accommodate circuit-switched and packet-switched traffic in a unified manner. The switch fabric is self-routeing and uses fixed-length minipackets within the switching fabric for all types of connections. Its kernel architecture is based on a routeing topology with individual connection paths from all inputs to all outputs and with FIFO queuing at each output. Owing to the disjoint connection paths, there is no internal blocking, and because of output queueing, output port blocking is prevented to a great extent. The uniformity in architecture allows construction of any size fabric from a single basic module which could be realized on a single chip. Larger-size configurations can be realized either as single-stage or multistage configuration. The second part of this paper discusses performance aspects and gives results and dimensioning guidelines for both circuit-switched and packet-switched traffic.     

CFD based evaluation of heat transfer coefficient from cylindrical particles

Two- and three-dimensional CFD modeling of heat transfer from discrete circular cylindrical particles in four different situations including A) infinite cylinder in cross-flow, B) cross-flow on finite cylinder with different aspect ratio in a rectangular duct, C) axial-flow on finite cylinder and D) axial-flow on finite cylinder with upstream turbulence have been investigated with the commercial CFD software, FEMLAB. The results were validated using experimental data from different research papers and also experimental correlations and show good quantitative and qualitative agreement with each other. In case B, a correction term has been proposed from CFD work which is applied to an experimental correlation to consider the aspect ratio influence on predicting the Nusselt number with an average error of 3.7%.     

Prediction of restrained shrinkage based on restraint factors in patching repair mortar

Shrinkage of repair material, especially in patching repairs, is the major factor inducing cracking in concrete repairs. Induced cracks in repair materials are due to restrained shrinkage. Although in usual practice, the free shrinkage of the repair mortar is measured, in reality, cracking is not due to free shrinkage. It is well known that cracking is due to restrained shrinkage. It is very hard to measure the restrained shrinkage; therefore, to overcome this problem a restraint factor (R) is used to modify the free shrinkage and come up with the restrained shrinkage. The restraint factor is influenced by the surface and boundary condition.In this study, the restraint factor for patching repair with different boundary conditions (with eaves and without eaves) and surface condition (rough and smooth) of the substrate concrete is investigated.The results show that the restraint factor R lies between 0.1 and 0.94; with an increase of restraint, the restraint factor is increased. In situations with a high level of restraint (eaves at the perimeter and a rough surface of substrate), the average R is 0.83. while with a low level of restraint (without eaves at the perimeter and a smooth surface of the substrate), the average R is 0.22.     

Optimization of STM/FIM nanotip aspect ratio based on the Taguchi method

This paper presents optimization of electrochemical etching parameters to achieve the optimum aspect ratio of the scanning tunneling microscopy/field ion microscopy tungsten nanotip by using Taguchi method. The combination of optimum level of process parameters was obtained by using the analysis of signal-to-noise ratio. The level of importance of the process parameters on the nanotip aspect ratio was determined by using analysis of variance. It was found that the optimum level of process parameters are electrolyte concentration of 2 M/lit, wire immersion length of 2.5 mm, cathode tube inner diameter of 40 mm, and voltage of 3.5 V. Within the range of experiments and the process parameters in terms of impact significance were found to be electrolyte concentration, process voltage, wire immersion length, and inner diameter of cathode tube, respectively. By using the optimum level of the process parameters, the nanotip aspect ratio was enhanced by 263% in comparison to the mean value of the experimental results. The nanotip aspect ratio of up to 163:1 was obtained in the present research.