NASHRIYYAH -I MUHANDISI -I BARQ VA MUHANDISI -I KAMPYUTAR -I IRAN, A- MUHANDISI -I BARQ
Year:2010 | Volume:8 | Issue:3|Papers Count:7

Blind estimation of number of active users and synchronization of those have important role in identification of system’s parameters, for the spectrum surveillance system design of multi-user direct sequence spread spectrum systems. For the number of active user estimation, signal’s eigenvalues of the received signal covariance matrix is distinguished from those of noise using an adaptive threshold. Analytical results and computer simulations show that signal of users with limited received power difference (less than 1.5 dB) can be synchronized using maximizing the Frobenious norm of the received signal covariance matrix. Moreover, this method is robust against of the carrier frequency offset which is due to Doppler shift in the carrier frequency of different users. By increasing difference among received powers of processed signals, performance of this method will be degraded. Hence, an iterative algorithm for interference cancellation of users with higher received power is proposed, which can be synchronize and estimate the number of active users with lower received power. The proposed method for number of active user estimation reduces computational complexity in comparison with the traditional methods for number of active user estimation, such as MDL and AIC, while keep the accuracy of estimation in low SNR regime. The performance of blind synchronization method in the flat fading channel is degraded; hence, a multi-antenna receiver is proposed to improve the probability of correct synchronization.

In this paper, a new method for identification of dominant source in creation of current and voltage distortions in PCC bus is presented. In proposed method, using superposition method along with definition of harmonic contribution and combining them with the critical impedance algorithm, new indices are defined and novel algorithm is presented. In this algorithm contrary to the critical impedance method, there is no need for values of customer's harmonic impedances. Also there is no probability that proposed algorithm lead to incorrect identification of dominant source even in cases in which the shares of PCC sides are close to each other. This fact is shown by comparison between proposed algorithm and critical impedance method. The analysis is verified on IEEE 13-bus test system using DigSilent software.

Bearing current in inverter-fed induction motors can result in premature bearing failure. This paper gives a detailed analysis of common mode voltage as the main source of bearing currents. Moreover, it presents an algorithm to reduce the common mode voltage in multilevel DTC drive. Based on the given analysis, the common mode voltage is reduced to about 33 percent of the initial value. The simulation results in MATLAB/SIMULINK confirm the capability of the algorithm in common mode voltage reduction without degrading the electromagnetic torque and the speed response. The results of real world experiments using TMS320F2812 DSP processor are also provided.

Faults in MV systems are the cause of almost 70 percent of outages in power system. By using suitable outage management methods, amount of power outages will be reduced in large extent. In Iran, traditionally outage management system was conducted by trial and error methods and relied on operators’ experience. The ever-increasing expansion of distribution networks has made it impossible to use traditional outage management systems. Outage management systems depend upon load type, load sensitivity, network type, amount of accumulation of load and population, and being rural or urban network.  In this paper, at first a technical assessment for implementing outage management methods in distribution systems of Iran is carried out. An economic assessment of implementing outage management system in urban distribution systems of Iran with dense population is conducted. For this purpose a sample network of Afsarieh area of Tehran has been chosen.

The redundancy problem of sensor response in electronic noses is still remarkable due to the cross-selectivity of chemical gas sensors which can degrade the classification performance. In such situations, a more efficient multiple classifier system can be obtained in random feature space rather than in the original one. Ensemble Feature Selection (EFS) methods assume that there is redundancy in the overall feature set and better performance can be achieved by choosing different subsets of input features for multiple classifiers. By combining these classifiers the higher recognition rate can be achieved. In this paper, we propose a feature subset selection method based on hierarchical clustering of transient features in order to enhance the classifier diversity and efficiency of learning algorithms. Our algorithm is tested on the UCI benchmark data sets and then used to design an odor recognition system. The experimental results of proposed method based on hierarchical clustering feature subset selection and multiple classifier system demonstrate the more efficient classification performance.

The elliptical structure for the micro electromechanical switches with low actuation voltage is proposed to work at the Ka band. The actuation voltage of this structure is shown to further been reduced by 21% with transverse cutting patterns. This Switch will then be analyzed to extract its parameters such as Insertion Loss, Return Loss, and Deformation Posture. The effect of the actuation voltage on the deformation of the bridge will then be analyzed and the results would be compared with other common bridges. This switch may be used as a low loss and effective element for much complicated systems such as distributed phase shifters and phased array applications.

Fuzzy clustering Method (FCM) is a commonly used method of data clustering. But, when too much data are available, the use of this method usually may lead to non-homogeneous distribution of data. In this paper a new method for clustering of wind speed data in wind farms is presented. In this method, using the PSO algorithm, wind speed data is clustered and the obtained results are compared with those of FCM and K-means clustering methods. Simulation results indicate the proposed method has better convergence than K-means and FCM methods, especially in conditions which too much data are not available.