INTERNATIONAL JOURNAL OF INDUSTRIAL ENGINEERING AND PRODUCTION MANAGEMENT (IJIE) (INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE) (PERSIAN)
Year:2000 | Volume:11 | Issue:4|Papers Count:12

The successful transmission of high speed digital data through any medium requires a knowledge of the degradation and distortions introduced into the received signal by the medium itself. This knowledge can then be used to design suitable detection schemes, usually based around adaptive algorithms .In the case of a frequency selective fading channels, these degradation are mainly caused by multipath propagation of the transmitted signal energy, which for the realistic cases of the non homogeneous media the received scattered signal will be time-varying in nature It follows That the received signal can suffer severe and rapid amplitude fluctuation (fades) across the bandwidth of the signal. In this paper hardware simulators with different Doppler spreads for the test of such a realistic data transmission systems are presented. These are then used to design an adaptive data transmission system over frequency selective fading channels.

In this paper, with the aid of Galerkin's method and integral equations of electric field, full wave analysis applies on wraparound microstrip antenna, which its result, is reach the parameters as input impedance, radiation frequency and radiation pattern. Also, in addition of the above result, a parallel feed network is designed, using the Tacheston software.

This paper aims to present a numerical method for solving the problem of multiple forward diffraction over an array of parallel rows of absorbing half - screens with arbitrary heights and spacings. The distances between half - screens are assumed to be much larger than the operating wavelength.The motive for conducting this research is the interest in prediction of macrocellular propagation of UHF radio waves in urban enviromnents and the cited absorbing half -screens approximate parallel rows of buildings. By exerting some modifications the scope of the presented model applications is expanded in such a way that primary model restrictions (namely the assumption of equi -height rows of uniformly spaced half -screens and bounding the geometry of the problem to the radiations which are beamed from over the rooftops) are eliminated.The obtained model is flexible enough to permit evaluation of the slow fading effects caused by variations in heights and spacings of rows of buildings. In this paper this capability has been made use of in order to check the validity of the results obtained by the proposed numerical method and good agreement has been observed with the measured values. It is believed that such results can be considered useful in design of cellular mobile radio (CMR) systems with elevated base station antennas.

Nowadays mental disorders and body diseases is diagnosed using recorded scalp, deep and cortical electrical potentials. In this paper a novel method is presented to estimate the deep potential of the brain based on DLM technique using surface electrodes. The simulated results show while using this method and moving into depth we will be able to distinct different sources of activities inside the brain with high resolution, which is impossible by other general procedures.Increasing the number of scalp electrodes improves the accuracy of this method.

Secure Communication is as old as Human civilization. Ciphers have been used in times for war and peace as well as in commerce and politics. In recent years the growing rate of information value has enormously increased the importance of its security and safety. Today encryption is considered as a branch of telecommunication science rather than a mere art. Speech security can be assumed a subject in the speech signal processing field. In this paper those methods have been received attention which are able to make information content of speech signal unintelligible for unauthorized use.Structure of the cited methods can basically be summarized as follows. In the first step the time domain samples of speech signal are transferred to a specific transformation domain and afterward the transformation coefficients are subject to a special permutation. Finally the encrypted speech signal in time domain will be obtained using the inverse transformation. It is obvious that the main difference between these methods is created by the type of exploited transformation. In this study a variety of transformation which have wide application in signal processing field have been investigated and their ability in encryption of speech signal under channel restrictions have been compared. Finally a modified trans formation based encryption scheme has been presented which effectively reduces the residual intelligibility of encrypted speech signal.

A method for designing the implicit generalized minimum variance self -tuning controllers is introduced so that, in addition to locating the desired poles for closed -loop system, a combination of the control and the output signal variances is also minimized.For this purpose, first, the orders of the controller and the weighting polynomials of the cost function are increased; then by applying the condition of the minimum combination of the control and the output signal variances, controller's coefficients are calculated provided that the pole locations of the closed - loop system are not changed. The simulation results show considerable reduction in variances without any change in the pole locations.

Harmonic state estimation in power system is allocation of data to amplitude and phase of voltage harmonics (state variables) in the buses of system by using measurements in the system and a special evaluation criteria. Measurements are voltage phasor and current phasor in the busbars and current phasor in the lines of system, and criteria of evaluation is based on weighted least squares method. Harmonic model in the estimation algorithm provide with attention to system topology and measurements achieved in the system. By classification of state variables in a few stages, dimension of model can be reduced to minimum level to increase speed and efficiency of estimation in the system. Results of simulation in the test system show accuracy and efficiency of this type of estimation and its ability to identify the harmonic sources in the system. Some of the characteristics of harmonic state estimation in power system is, it's enforceable with Kalman filter algorithm, it's enforceable by using neural Network, and it's enforceable in unbalanced three phase power systems.

In this paper the output port of an industrial automation system which includes Data Registers, Digital to Analog Converters, Decoders and Buffers is integrated using CMOS Technology. The designed circuits have been simulated in SPICE using 1.5 micrometer CMOS Technology. The designed output port is eight bits, with -5 V power supply, which is capable of handling data with a frequency of 5 :MHz with a power consumption of 29.3 mw. The fractional Temperature coefficient of the current reference is very low and in the rage of 333 PPM/oC. The settling time for the full scale input is about 700 nsec.s and the output voltage swing is from zero to 4.3 volts.

The paper presents theory of designing for DCM dynamic stablizer in HVDC-AC system . First and second order of eignvalue sensitivity method is used for designing. Dynamic model which is obtained in the paper is a new model. This model is based on reference from theory and linearization method .It is structured in such manner to enable study of HVDC system dynamic and analysis of interaction between this system with other AC network. Finally, results of implementing of implementing of this DCM modulator on a typical network is compared with the performance of conventional PSS dynamic stablizer.

The use of a superconducting magnetic energy storage unit for load leveling/damping purpose is presented in this paper. The state variable equations representing the SMES unit and a block diagram defining the input-output structure is developed for it to study the power system performance using a digital computer. A digital computer simulation model of the 2 - area power generation systems is developed for dynamics and stability evaluation.Sudden small variation in electric power load occurred in every time and continuously disturb the normal operation of a power load occurred in every time and continuously disturb the normal operation of a power system. The crux of the problem lies in the fact that the inertia of the rotating parts is the only energy storage capacity in a power system. Thus, following a sudden increase of load, the only source from where the immediate requirement of additional power can be met is the kinetic energy in the generator rotor, This imposes variation in frequency . The problem is even more serious with the modern generator (and futuristic superconducting generators ) which are designed to have low inertia in comparison with their power rating. increasing the rotor inertia by flywheel attachment on the rotor shaft is not always acceptable due to the severe tortional stress developed during dynamic oscillations. The addition of a fast acting energy storage unit can improve the performance of the system under such conditions. This paper presents briefly some of the important application of superconducting magnetic energy storage unit to improve dynamic performance of the power system. The application of superconducting magnetic energy storage to the stabilization of a power system with long distance bulk power transmission lines which has the problem of poorly damped power oscillation is presented. Control scheme for stabilization using SMES are discussed in details and a proportional integral damping controller designed by pole assignment method.

Methods developed for nonlinear system identification, usually involve complicated and long calculations. In recent years neural networks has received considerable attentions for nonlinear dynamic system identification. A multilayer perception (MLP) neural network< with back- propagation learning algorithm is usually used in different papers for identification of simple nonlinear systems.In this paper, first it has been shown that such algorithm fails to identify a highly nonlinear and high order system such as a turbogenerator. Then it has been shown that the proposed structure can well identify the system when the marquardt learning algorithm is used.The ultimate goal of an identification algorithm is parallel identification, that is, using the output of the model instead of the output of the system in the learning algorithm. This is also achieved with the proposed structure and learning algorithm when some state variables of the system is used for system identification.Simulation results shows the validity of the identified model even when the operating conditions of the system is changed.

In this paper, squirrel cage induction motor is modeled by magnetic equivalent circuit approach under broken rotor bar condition.Then detection of broken rotor bar is studied and discussed by applyinga new method. In this new approach, induction motor is detached from the three-phase supply. After removal of supply, stator currents are zero and induced voltages in stator windings due to rotor flux is measured.Only rotor currents produce rotor flux and if there is any broken bar, it will affect the induced voltages in stator windings. Using frequency spectrum and envelope of induced voltages, broken rotor bares) can be detected. Due to supply removal, there is no effect of source time harmonics in the detection of broken bars and due to flux reduction, nonlinearity of machine magnetizing characteristics due to saturation can be avoided.