Scientia Iranica
Year:2014 | Volume:21 | Issue:3 (TRANSACTIONS D: COMPUTER SCIENCE AND ENGINEERING AND ELECTRICAL ENGINEERING)|Papers Count:14

Since there is no dierence between uplink and downlink subframes in the IEEE 802.16 mesh mode, both downlink and uplink tra cs are transmitted within a single time frame. Hence, in most scheduling methods, only one scheduling algorithm is used for both uplink and downlink tra c. However, because of the dierent characteristics of uplink and downlink tra c, dierent scheduling methods should be used for each of the tra c types. In this paper, we focus on the mesh centralized scheduling of downlink ow in the data subframe. After comprehensive analysis of the characteristics of downlink tra c, we propose a new algorithm, called Tra c-Aware Scheduling (TAS). In this algorithm, the downlink tra c distribution is tuned for maximum concurrent transmission rate. This goal is achieved by choosing dierent link selection criteria, such as maximum demandant sender, farthest receiver and least interfered path, based on the downlink tra c analysis results. The simulation results show that our algorithm outperforms existing methods in terms of scheduling length, link concurrency, and throughput (about 13.7% in average) for the downlink tra c. Moreover, the proposed algorithm is scalable. In particular, on average, 3% improvement is achieved in terms of scheduling length at higher tra c loads and in the number of nodes.

Recently, the strategy for forecasting option price has become a popular nancial topic because options are important tools on risk management in nancial investments. The well-known Black-Scholes model (B-S model) is widely used for option pricing. In B-S model, the normal distribution assumption is important. However, the nancial data in real life may not follow the normal distribution assumption. For this reason, this paper proposes a novel hybrid model, which is a nonlinear prediction model without normal distribution assumptions to forecast the option prices. The proposed model is composed of a Fuzzy Time Series (FTS) model, a Least Square Support Vector Regression (LSSVR), and a bootstrap method. In the proposed model, FTS model is rstly used to fuzzify dataset and to build historical database. Subsequently, the proposed method uses the remainder contractual time to search similar historical trends as training samples. Finally, we use the bootstrap method on LSSVR to enhance the prediction accuracy. The experiment results show that the proposed model outperforms traditional time series models and several hybrid models in terms of the Root Mean Square Error (RMSE), the Mean Absolute Error (MAE) and the correlation coe cient (r) of actual and forecasted option price.

ASIC implementation of a high speed multiplier using a high accuracy oating point logarithmic number system is reported in this paper. The most popularly used techniques for computing logarithmic calculations for digital signal processors are: Lookup table based implementation, polynomial approximation, and Taylor series expansion. But, all these techniques suer from low accuracy, due to the choice of having only lower order terms of the expanded series. In the present work, logarithmic conversion is implemented by a oating point (IEEE-754 single precision) converting methodology, thereby eliminating series expansion, which eventually results in high accuracy. The improvement in speed, by avoidance of carry propagation, was achieved through Canonical Signed Digit Code (CSDC) implementation, while the high accuracy was achieved through an error minimization circuitry especially designed for this purpose. The functionality of these circuits was checked, and performance parameters, like propagation delay and dynamic power consumption, were calculated by spice spectre using 90 nm CMOS technology. The propagation delay and power consumption of the resulting (128 128) bit multiplier (divider) was only 93 ns and ~ 80 mW, respectively, for a layout area of ~ 25mm2. This implementation oered a signi can’t improvement in terms of accuracy, delay and power from those reported earlier.

A new method for increasing the speed of all-optical Mach-Zehnder Interferometric (MZI) switching with a bulk Semiconductor Optical Ampli er (SOA), using chirped control signals, is suggested and theoretically analyzed. For 125fs input and chirped control pulses, we show acceleration of the gain recovery process using the cross phase modulation (XPM) eect. Our method depicts that Tb/s switching speeds, using bulk SOA-MZI with a proper Q-factor, is feasible. For the rst time, we reach operation capability at 2Tb/s with a Q-factor of more than 10. The new scheme also improves the extinction and amplitude ratio of the output power, as well as increasing the contrast ratio of the switched signal. We use a nite dierence beam propagation method for MZI analysis, taking into account all nonlinear eects of SOA, such as Group Velocity Dispersion (GVD), Kerr eect, Two Photon Absorption (TPA), Carrier Heating (CH) and Spectral Hole Burning (SHB).

In this paper, a wideband coaxial power combiner with tray con guration, employing twenty 1-Watt elements mounted on antipodal nline boards, is implemented to achieve a rather high power microwave ampli er in the frequency band of 6-18 GHz. The average gain, Continuous Wave (CW) output power and Power-Added E ciency (PAE) of the combiner are measured as 17 dB, 15 Watts and 17%, respectively with ripples of less than 3 dB. In order to investigate the combining e ciency, the passive structure is fabricated and tested in the rst step. Its main components, i.e. the coaxial adapter and the antipodal nline boards, are synthesized by using Klopfenstein's optimum impedance taper, and simulated by Ansoft HFSS. Next, the active devices are added to the system. Stability, thermal management and life time issues are examined and addressed in this step. The system reliability is also demonstrated as graceful degradation due to the failure of devices.

This paper obtains the soliton and other solutions to a few nonlinear wave equations that arise on a daily basis in various engineering disciplines and other fields. The sine-cosine method is adopted to extract these solutions. The ansatz method is also implemented to obtain a singular soliton solution to the Schrodinger-Hirota equation that is studied in electrical engineering in the context of nonlinear ber optics. In this context, both Kerr law and power law nonlinearity are going to be addressed. There are several constraint conditions that will be listed in order for the solutions to exist.

This paper presents a Teaching-Learning-Based Algorithm (TLBO) to solve Economic Load Dispatch (ELD) problems involving dierent linear and non-linear constraints. The problem formulation also considers non-convex objective functions including the eect of valve-point loading and the multi-fuel option of large-scale thermal plants. Many diculties, such as multimodality, dimensionality and dierentiability, are associated with the optimization of large scale non-linear constraint based non-convex economic load dispatch problems. TLBO is a population-based technique which implements a group of solutions to proceed to the optimum solution. TLBO uses two dierent phases, ' Teacher Phase' and ' Learner Phase', and uses the mean value of the population to update the solution. Unlike other optimization techniques, TLBO does not require any parameter to be tuned, thus, making its implementation simpler. TLBO uses the best solution of the iteration to change the existing solution in the population, thereby increasing the convergence rate. In the present paper, Teaching-Learning-Based Optimization (TLBO) is applied to solve such types of complicated problems eciently and eectively, in order to achieve a superior quality solution in a computationally ecient way. Simulation results show that the proposed approach outperforms several existing optimization techniques. Results also proved the robustness of the proposed methodology.

Compared to multistage grid-connected photovoltaic systems (GCPVSs), complexity in monitoring and controlling scheme in a single-stage GCPVS is much increased, as maximum power point tracking and power sharing with the grid need to be considered simultaneously in the same stage. An e cient technique for performing the aforementioned functions, simultaneously, in a single stage system, by adjustment of the voltage phasor of a voltage source inverter, is presented. An active and reactive power ow from the inverter is regulated by the adjustment of amplitude and phase angle of the inverter output voltage with respect to the grid voltage.An algorithm is also developed for automatic active and reactive powers sharing for the intended applications in the single stage GCPVSs. A typical 2.8 kWp, 368 V (DC) nominal photovoltaic array based GCPVS is modeled and simulated to analyze proper load matching for dierent ambient and grid voltage conditions. Performances, like the e ciency of the inverter and the total harmonic distortion of the inverter output voltage and current, are studied, with variations in ambient conditions and the modulation index of a sinusoidal pulse width modulation of the IGBT based inverter. Simulation results have been included to show the feasibility of the proposed technique. ciency of the inverter and the total harmonic distortion of the inverter output voltage and current, are studied, with variations in ambient conditions and the modulation index of a sinusoidal pulse width modulation of the IGBT based inverter. Simulation results have been included to show the feasibility of the proposed technique.

According to dynamic core hypothesis, integration and dierentiation are main properties of consciousness. Hence, we expected that the consciousness neuronal correlate covers these properties in structural level. These properties could be captured in smallworldness properties, i.e. high clustering coe cient and low path-length. Thalamocortical (TC) loop and cortex are two main candidates for Neural Correlates of Consciousness (NCC). We studied small-worldness in these systems. For this purpose, we calculated clustering coe cients, characteristic path lengths and their robustness against lesions. We simulated lesions in two ways: eliminating connections, and deleting nodes. We used anatomical connections of TC and cortex of macaque from the CoCoMac neuroinformatic database. Our results show that: 1) Lesions causes an increase in path length and decrease in clustering coe cient which cause the destruction of the integration and segregation capabilities of brain network, 2) Deleting the connections is more destructive than deleting the nodes, 3) During high levels of lesions, the thalamo-cortical connections are more important than cortico-cortical connections in the sense of clustering coe cient. In terms of path-length, during high levels of nodes' lesions, the thalamo-cortical connections are more important than cortico-cortical connections, while during edges' lesions cortico-cortical connections are more important than thalamo-cortical connections.

A transient phenomenon that leads to the false trip of the power transformer dierential protection during the energization of a loaded power transformer is the ultra-saturation phenomenon. In this paper, at rst, a new algorithm for three-phase power transformer dierential protection, considering the eect of the ultra-saturation phenomenon, is presented. To model the ultra-saturation phenomenon, the nonlinear characteristic of the transformer core and the eect of saturation of the current transformers are taken into account. It is assumed that the load of the transformer is resistive and inductive. In this algorithm, the ultra-saturation phenomenon, external and internal faults of the power transformer and magnetic inrush current are simulated, and appropriate criteria using the signal harmonic components of the dierential current will be presented for the Discrete Fourier Transform (DFT) algorithm to distinguish between. In this paper, simulation is done by PSCAD and MATLAB programs.

Recent advancements in photovoltaic (PV) system technologies have decreased their investment cost and enabled the construction of large PV farms for bulk power generations. The output power of PV farms is aected by both failure of composed components and solar radiation variability. These two factors cause the output power of PV farms be random and dierent from that of conventional units. Therefore, suitable models and methods should be developed to assess dierent aspects of PV farms integration into power systems, particularly from the system reliability viewpoint. In this context a reliability model has been developed for PV farms with considering both the uncertainties associated with solar radiation and components outages. The proposed model represents a PV farm by a multi-state generating unit which is suitable for the generation system assessment. Utilizing the developed reliability model, an analytical method is proposed for adequacy assessment of power generation systems including large PV farms. Real solar radiation data is used from Jask region in Iran which are utilized in the studies performed on the RBTS and the IEEE-RTS. Several dierent analyses are conducted to analyze the reliability impacts of PV farms integration and to estimate the capacity value of large PV farms in power generation systems.

In this paper experimental and analytical explorations of an R-2L-2D circuit were carried out. The responses from the ordinary em RL-Diode and R-2L-2D circuits were characterized and compared for a wide parameter region. As a new circuit, R-2L-2D has an additional inductor and a diode. The circuit had dierent attractors compared with the ordinary RL-Diode circuit. It was proven that the new circuit exhibited wider chaotic regions on the parameter space (i.e. input voltage, Vf, and frequency, f). Both even and odd subharmonic responses were observed following the multiple periodic doublings. The bifurcation analysis revealed the dominance of feeding frequency by means of the center manifold theory. However, periodic and chaotic attractors diered for each circuit. In fact, the new circuit generated symmetric trajectories. A detailed investigation proved that the chaotic responses in the proposed circuit could start at the peak-to-peak voltage of Vf=1: 35 V, at frequency 40 kHz, which was nearly half of the frequency value found for the ordinary circuit. Besides, a wide range of chaotic behavior was observed beyond Vf=0: 675 V and f=200 kHz. Chaotic trajectories dominated the dynamics up to f=500 kHz.

The analysis and design of a tunable low noise active inductor is presented The noise performance of the proposed gyrator-based active inductor is improved without either degrading its quality factor or consuming more power using a linear Feed Forward Path (FFP). The proposed low noise active inductor has been designed and fabricated using standard 0.18-mm CMOS technology. The measurements show a 3 fold improvement in the input noise current compared to that of conventional active inductors. The active inductor was tuned and measured at the resonance frequency of 2.5 GHz, which could be extended as high as 5.5 GHz, with a quality factor of 30. The circuit draws 4.8 mA from a 1.8 V supply.

This paper presents analysis of the Static Noise Margin (SNM), power dissipation, access time and dynamic noise margin of a novel low power proposed 8T Static Random Access Memory (SRAM) cell for read operations. In the proposed structure, two voltage sources are used, one is connected with the bit line and the other is connected with the bitbar line in order to reduce the voltage swing at the output nodes of the bit and the bit bar lines. Simulation results for the read static noise margin, read power dissipation, read access time and dynamic noise margin have been compared to those of other SRAM cells, reported in dierent literatures. It is shown that the proposed SRAM cell has better static noise margin and dissipates less power in comparison to other SRAM cells. Analog and schematic simulations have been done in a 45 nm environment with the help of Microwind 3.1, using the BSimM4 model.