Scientia Iranica
Volume:19 Issue: 3, 2012

Stream programming languages have become popular owing to their representations that enable parallelization of applications via static analysis. Several research groups have proposed approaches to software pipeline streaming applications onto multi/many-core architectures, such as CELL BE processors and NVIDIA GPUs. In this paper, we present a novel scheduling algorithm that software-pipelines streaming applications onto multi/many core architectures. The algorithm generates software pipeline schedules by formulating and solving MILP (Mixed Integer Linear Programming) problems. Experimental results show that compared to previous works, our approach generates schedules that use up to a 71% smaller amount of buffers needed for communication between kernels.

Stream programming languages have become popular owing to their representations that enable parallelization of applications via static analysis. Several research groups have proposed approaches to software pipeline streaming applications onto multi/many-core architectures, such as CELL BE processors and NVIDIA GPUs. In this paper, we present a novel scheduling algorithm that software-pipelines streaming applications onto multi/many core architectures. The algorithm generates software pipeline schedules by formulating and solving MILP (Mixed Integer Linear Programming) problems. Experimental results show that compared to previous works, our approach generates schedules that use up to a 71% smaller amount of buffers needed for communication between kernels.

The tensor based classifier has attracted a great deal of interest, due to its representation of input objects in a natural form in overcoming small sample size problems and in providing high classification accuracy. Multi-Linear Discriminant Analysis (MLDA) is an efficient classifier, which employs tensor properties to simplify computation and improve accuracy. In this study, a boosting framework is exploited to further improve a tensor-based MLDA classifier. In the boosting framework, several weak learners are trained with different distribution of training samples and, then, integrated with suitable weights to build a strong classifier with a high generalization capacity. In our proposed method, namely BMLDA (Boosted MLDA), the MLDA classifiers are weakened and considered as feature projection components (weak learners) in the boosting framework. Finally, a Nearest Neighbor (NN) classifier makes the final decision and enables the BMLDA to act as a multi-class classifier. To assess BMLDA, several versions of Linear Discriminant Analysis (LDA) classifiers, such as boosted direct LDA, direct LDA, subclass-LDA, MLDA and LDA, were implemented. Empirical evaluations on two real applications demonstrated that the proposed BMLDA outperformed its competitors. The proposed method is beneficial in exploiting the boosting framework to accommodate tensor-based learners that totally construct a powerful multi-class ensemble classifier with much higher performance.

This study investigates different types of career orientation of a sample of Iranian software engineers and antecedents of these orientations. A qualitative study was conducted in seven Iranian small to large sized companies, where forty-nine software engineers were interviewed. Using the thematic analysis technique, technical, managerial, entrepreneurial, project based, and hybrid orientations were identified. For some orientations, several sub-orientations were also identified. In addition, the results propose some antecedents of career orientation, mainly based on engineer needs, competencies and values, which may be moderated by some external factors, including organizational and national level phenomena.

Tzeng et al. proposed a threshold multi-proxy multi-signature scheme with threshold verification. Recently, Hsu et al. pointed out that Tzeng et al.’s scheme was vulnerable to insider attacks and proposed an improvement to eliminate the pointed out security leak. We will show that Hsu et al.’s improvement cannot resist the frame attack. That is, after intercepting a valid proxy signature, an adversary can change the original signers to himself, and forge a proxy signature. To remedy this weakness, we will propose a new method.

The advent of Cloud computing as a new model of service provisioning in distributed systems, encourages researchers to investigate its benefits and drawbacks in executing scientific applications such as workflows. In this model, the users request for available services according to their desired Quality of Service, and they are charged on a pay-per-use basis. One of the most challenging problems in Clouds is workflow scheduling, i.e., the problem of satisfying the QoS of the user as well as minimizing the cost of workflow execution. In this paper, we propose a new QoS-based workflow scheduling algorithm based on a novel concept called Partial Critical Paths (PCP), which tries to minimize the cost of workflow execution while meeting a user-defined deadline. This algorithm recursively schedules the partial critical paths ending at previously scheduled tasks. The simulation results show that the performance of our algorithm is very promising.

With increasing use of data transmission between digital systems as well as subsystems, the need for more reliable communication is felt. In this research, a new approach to the linear error correcting codes is introduced. Among advantages of this technique are simpler code construction and also decoding algorithm, as well as easier understanding of the basic concept. Based on an earlier paper and also recent work done, we first discuss briefly the use of Karnaugh map and its advantages in constructing simple codes. We then prove a theorem on application of the map to multiple error correcting codes. Using these results, we propose a simple technique that leads to obtaining a code with more capabilities. We also discuss another advantage of using the Karnaugh map in error correcting codes. As an example, we discuss a special case of generating a double error correcting code and using Karnaugh map features to give it extended capabilities.

Quality of Experience (QoE) enhancement for a transmitted video sequence in Mobile Ad hoc NETworks (MANETs) is a challenging and important issue in the networking research community. Intrinsic high levels of packet error probability in MANETs can cause high levels of distortion based on the position of packet loss in transmitted frames. Hence, exact modeling of the impact of packet loss on video quality and the resulting distortion is an important task. Many conventional distortion modeling techniques simply consider a linear relationship between the packet loss and distortion, which is inaccurate. The main contributions of the current work are twofold. At first, an accurate model has been developed, which can capture the exact effect of network packet loss on video quality performance (and hence on the QoE) with Group Of Picture (GOP)-level granularity. Then, based on this model, an optimal bandwidth allocation strategy is developed in MANETs, with which, based on some network-specific constraints, the loss-induced distortion associated with a video source is minimized, using some cross-layer design techniques. Finally, the resulting optimal rates can be used as rate-feedbacks for on-line rate adaptation of an appropriate video encoder, such as H.264/AVC. Numerical analysis verifies the theoretical results.

This paper proposes a novel method for an optimal cabling route; a combinatorial design algorithm through heuristic mathematical formulations to choose the optimal path when two or more physical paths are available. The best path, from an EMC point of view, can be chosen based upon various criteria, such as monetary cost minimization, voltage drop and quality (EMC) parameters. This paper initially provides numerical 2D and 3D resolutions of the problems of radiation generated by current sources. The proposed approach is based on the finite element method, associated with absorbing boundary conditions. The presented model makes it possible to consider wave propagation and its effect on heterogeneous mediums in transient, which can be applied in EMC for simulation of radiation. Initially, the formulations for electromagnetic modeling of the proposed problem are specified in detail. The simulation results are used for the cabling routes, with respect to undesirable field distribution in the specified regions. Based upon the aforementioned criteria, using an Analytical Hierarchy Process, the intelligent choice of optimal cabling routes can be achieved. Hence, there is a need for the proposed method to be successfully implemented on two different types of power system; a current injection system, and high voltage substation.

A method based on transforms and support vector machines was presented for fault diagnosis of power electronics circuits in which the transform time-frequency analysis of the fault signal is used to extract the features corresponding to various faults. Then, fault types are identified through the pattern recognition classifier, based on SVM. The simulation results show that the proposed method can accurately diagnose faults and locate fault elements for power electronics circuits. It also has excellent performance for noise robustness and calculation complexity, thus, having good practical engineering value in the solution to fault problems for power electronics circuits.

This paper deals with the design of steerable concentric rings array for low side lobe level using evolutionary optimization. This design of steerable concentric rings arrays considers the optimization of the amplitude and phase excitations across the antenna elements by using genetic algorithms. Some results obtained by simulation for a steerable concentric rings array are presented. The results shown in this paper present a good performance in the array factor response for the optimized design with evolutionary optimization.

This study compares the dynamic behaviors of the Lorenz system with complex variables to that of the standard Lorenz system involving real variables. Different methodologies, including the Lyapunov Exponents spectrum, the bifurcation diagram, the first return map to the Poincaré section and topological entropy, were used to investigate and compare the behaviors of these two systems. The results show that expressing the Lorenz system in terms of complex variables leads to more distinguished behaviors, which could not be achieved in the Lorenz system with real variables, such as quasi-periodic and hyper-chaotic behaviors.

Clipping and filtering is a well-known technique used at the transmitter side of Orthogonal Frequency Division Multiplexing (OFDM) systems to reduce the Peak to Average Power Ratio (PAPR). In this paper, the extension of clipping and filtering to spatially multiplexed OFDM systems will be introduced. In the proposed scheme, the channel coding and clipping operations are performed at the transmitter branches independently. Based on this scheme, a receiver is proposed, which is based on the detection method of Nulling Cancellation (NC). In this method, the layer with the minimum average noise power over all subcarriers is selected and an iterative method is used to compensate for its clipping noise. Then, the effect of the signal of the selected antenna is eliminated from the received vectors.

This paper presents a fuzzy-based genetic algorithm to maximize total system social welfare by best the placement and sizing of TCSC and SSSC devices, considering their investment cost in a double-sided auction market. To introduce more accurate modeling, the valve loading effects are incorporated into the conventional quadratic smooth generator cost curves. In addition, quadratic consumer benefit functions are integrated into the objective function to guarantee that locational marginal prices charged at the demand buses are less than, or equal to, the DisCos benefit, earned by selling the power to retail customers. The proposed approach utilizes fuzzy-based genetic algorithms for optimal scheduling of GenCos and DisCos, as well as optimal placement and sizing of SSSC and TCSC units. In addition, the Newton–Raphson approach is used to minimize the mismatch of the power flow equation. Simulation results on the modified IEEE 14-bus and IEEE 30-bus test systems (with/without line flow constraints, before and after the compensation) are used to examine the impact of SSSC and TCSC on total system social welfare improvement versus their cost. To validate the accuracy of the proposed method, several case studies are presented and simulation results are compared with those generated by genetic and Sequential Quadratic Programming (SQP) approaches.

In this paper, a new analytical technique, called the Optimal Homotopy Perturbation Method (OHPM), is suggested to solve a class of nonlinear Optimal Control Problems (OCP’s). Applying the OHPM to a nonlinear OCP, the nonlinear Two-Point Boundary Value Problem (TPBVP), derived from the Pontryagin’s maximum principle, is transformed into a sequence of linear time-invariant TPBVP’s. Solving the latter problems in a recursive manner provides the optimal trajectory and the optimal control law, in the form of rapid convergent series. Furthermore, the convergence of obtained series is controlled through a number of auxiliary functions involving a number of constants, which are optimally determined. In this study, an efficient algorithm is also presented, which has low computational complexity and fast convergence rate. Just a few iterations are required to find a suboptimal trajectory-control pair for the nonlinear OCP. The results not only demonstrate the efficiency, simplicity and high accuracy of the suggested approach, but also indicate its effectiveness in practical use.

Cognitive Radio (CR) is a promised solution to the lack of spectrum, as well as spectrum inefficiency, in current communication networks. A major demand of this technology is for fair spectrum assignment to unlicensed (secondary) users. This paper addresses a new approach to this challenge in CR networks, based on the Graph Coloring Problem (GCP) and the Ant Colony System (ACS). In addition, we compare the performance of the proposed algorithm with Color Sensitive Graph Coloring (CSGC) and Particle Swarm Optimization (PSO) based techniques.

The problem of gyrocompassing using inertial sensors, i.e., gyros and accelerometers, is addressed. North finding, with an order of accuracy of one arc-min, is not only required for the initial alignment of inertial navigation systems, but also has a critical role to play in the guidance and navigation of ships that navigate for long periods of time. In this work, after extracting the error model of an inertial navigation system and augmenting it with the error model of inertial sensors, a processing algorithm based on the Kalman filter is designed and simulated to process the navigation system velocity error, and to estimate and correct tilt and heading errors along with gyro drifts and accelerometer biases. It is verified that using gyros with drift stability of 0.01 deg/hr, and accelerometers with bias stability of 100 μg, the true north direction can be determined with an accuracy of about five arc-mins.

Forward Error Correction (FEC) codes may be used to protect a video code stream against packet erasures or errors when passing through an error-prone network. To have maximum possible fidelity at the decoder side, an Unequal Loss Protection (ULP) approach should be used to packetize the scalable video code stream, so that the different parts of the scalable video stream are protected according to their importance. Unequal loss-protected packetization leads to segmentation of the scalable code stream, such that the source can be reconstructed with the maximum possible fidelity at the decoder side. In Ardestani et al. (2009) [1] we have found an analytical relation between the optimal sizes of any two consecutive segments. This idea yields an efficient (as efficient as the local search algorithm in Stankovic et al. (2002) [2]) low-complexity progressive solution for the segmentation problem. In this paper, we use a progressive approach for ULP packetization of a scalable video stream generated from a T+2D encoder. In addition, an optimal rate allocation is used for optimal rate budget division between successive Groups Of Pictures (GOPs) of the video sequence. The experimental results demonstrate that the optimal rate budget allocation outperforms the conventional strategy of equal rate budget distribution up to 0.65 dB.

In this paper, new and efficient algorithms for solving optimal control problems and the controlled Duffing oscillator are presented. The solution is based on state parameterization, such that the state variable can be considered as a linear combination of Chebyshev polynomials with unknown coefficients. First, an optimization problem in (n+1)-dimensional space is changed into a one-dimensional optimization problem, which can then be solved easily. By these algorithms, the control and state variables can be approximated as a function of time. Convergence of the algorithms is proved and some illustrative examples are presented to show the efficiency and reliability of the presented method.

The interband transition rate and surface recombination rate of carriers in quantum dots, as two effective parameters to optimize the photocurrent and effciency of the intermediated band solar cells, have been classically studied. Formulation of these rates shows that they depend on the recombination lifetime of the carriers. This dependency may play the role of recombination or generation centers to the quantum dots. We have calculated these rates for two different values of recombination lifetimes. We have concluded that for a longer lifetime, quantum dots act as carrier generation centers and enhance the photocurrent and efficiency of the solar cell. It is also shown that there is an optimal number of stacked QD layers to be incorporated in the i-region in order to produce the maximum photocurrent.

This paper represents a developed procedure for the evaluation of flashover rates caused by both lightning vertical and non-vertical strokes on transmission lines, using a Monte Carlo method. The main goal of this paper is to increase the accuracy flashover rates through the accurate modelling of network components, such as footing impedance, the transmission tower and chain of insulators, and also simulation of lightning non-vertical strokes. Modelling network components, statistical parameters related to lightning and a problem-solving algorithm are described. Parametric studies using this procedure can also be performed to determine the sensitivity of the flashover rate with respect to some parameters of the return stroke and the transmission line. Simulation results obtained using MATLAB and the ATP/EMTP show that considering the lightning non-vertical instead of vertical strokes leads to a better evaluation of the lightning performance of transmission lines.

In recent years, distributed electrical power generation systems have attracted more attention due to their small scale technologies, lower cost electricity generation and higher reliability and security with fewer environmental consequences, rather than traditional power generators. Different applications such as peak saving, cogeneration, remote power and premium power will make it more and more useful. On the other hand, the permanent magnet synchronous generators, driven by micro turbines, play an important role in the distributed power generation systems. These PM generators produce a high frequency electrical power which needs to be converted to the usual power frequency, 50 or 60 Hz. This conversion is usually done using a back to back combination of rectifier, DC-link and inverter. The main focus of this paper is to present a model for a voltage source rectifier which is connected to a PM generator in a micro turbine unit, where Virtual Flux based Direct Power Control strategy is used to control the mentioned rectifier. The simulation results, using Matlab/Simulink, are presented to show the validity of the proposed model for the PWM rectifier, and to evaluate the performance of the control strategy.

In this paper, the operational Tau method is employed to approximate the solution of two dimensional linear Volterra integral equations of the first kind (2DLVIEF). To this end, we convert the 2DLVIEF to the equivalent equation of the second kind, and then we use the operational approach of the Tau method, with standard base, to convert it to the system of algebraic equations. Finally, some numerical examples are given to clarify the efficiency and accuracy of the presented method.

Many experimental and numerical studies were devoted to Dielectric Barrier Discharge (DBD) in air, but no mathematical models were proposed, either for current or for power. As they depend on several parameters, it is difficult to find a formula that considers many factors. The aim of this paper is first to make a brief comparison between surface and volume DBD, and second to model the current and power of a DBD in a “multipoints-plane” electrode system, using the methodology of experimental design. Three factors were considered: inter-electrode distance, distance between adjacent points, and thickness of the glass dielectric barrier. A double Composite Centred Faces experimental design (CCF) was carried out. The obtained results made it possible to propose mathematical models and, therefore to study interactions between various factors.