فهرست مطالب

  • Volume:3 Issue: 2, 2009
  • تاریخ انتشار: 1388/06/20
  • تعداد عناوین: 8
|
|
  • Maryam Shakiba, Mohammad Teshnelab, Sadan Zokaei Page 1
    Prediction is an important issue in many dynamical systems and is vital for effective management and control of plants. An important process which has recently derived much attention is the congestion control problem. Prediction of different traffic parameters can help in managing a congestion in a computer network. In this thesis, using real data for from the router between Iran Telecommunication Research Center and data Data company during December, January, February and March 2007, the router interval traffic rates are analyzed. Also, a comparative study is performed using the different methods employed and prediction results are provided to show the effectiveness of the predictions.
  • Sayedmasoud Hashemi Amroabadi, Mohammadreza Ahmadzadeh, Ali Hekmatnia Page 7
    Breast cancer is one of the leading causes of deaths among women. Mammography is currently the best method for early detection. Due to the breast tissue type and different kinds of lesions, by using low dose x-ray in mammography, the detection of lesions in mammograms becomes very ambiguous and a tedious work. Early detection is the most effective ways to reduce the mortality rate. Our main aim in this paper is detection and recognition of tumors in digital mammograms. Mammograms usually have a large size so the processing of the entire mammogram takes a lot of time. To reduce the size and therefore the processing time and also to decrease the False Positive Rate, a two-step algorithm is used. At the first step some unimportant regions such as background and pectoral muscle are eliminated and at the second step an ROI detection algorithm is proposed which extracts the most likely regions to tumors. To recognize the tumors in the detected regions, some features are extracted from each region. To find the most effective features for tumor detection, several data mining feature extraction and feature selection methods are used and then compared. To increase the performance and reduce the number of features a GA based algorithm is proposed. Finally, SVM is used as our classifier, because it has the best results in comparison with other tools in our application. Experimental results show that the performance of proposed methods is better than other previous methods. The True Positive Rate using SVM is 94.59% and the False Positive Rate is 22.95%.
  • Mohamadnaser Moghadasi, Zahara Ahangari Page 19
    Gate Induced Drain Leakage (GIDL) current is one of the main leakage current components in Silicon on Insulator (SOI) MOSFET structure and plays an important role in the data retention time of DRAM cells. GIDL can dominate the drain leakage current at zero bias and will limit the scalability of the structure for low power applications. In this paper we propose a novel technique for reducing GIDL and hence off-state current in the nanoscale single gate SOI MOSFET structure. The proposed structure employs asymmetric gate oxide thickness which can reduce GIDL current and hence Ioff current to about 98% in comparison with the symmetric gate oxide thickness structure, without sacrificing the driving current and losing gate control over the channel. This technique is very simple in the fabrication point of view in CMOS technology.
  • Ghazanfar Shahgholiyan, Ebrahim Haghjou, Saeed Abazari Page 25
    This paper presents a study about of a fuzzy Controlled STATCOM, which can be applied for mitigation of the voltage flicker in a distribution system. The voltage flicker is produced by a large variable load absorbing continuously changing currents such as an arc furnace. The DSTATCOM includes a voltage-sourced PWM inverter and its control system. The control strategy of the DSTATCOM plays an important role in maintaining the voltage flicker. Here, the DSTATCOM controller is designed with two types of controllers, linear proportional-integral (PI) and nonlinear fuzzy logic. The simulation of the DSTATCOM with 3MVar reactive power on a 25 KV distribution network is carried out in MATLAB/SIMULINK software. Finaly, fuzzy controllers were evaluated by comparing its performance with the PI controllers. It is observed that the fuzzy controllers are very superior in mitigating the voltage flicker.
  • Ebrahim Nasr Esfahani, Saeed Abazari, Cholamreza Arab Page 37
    In this paper a neuro-fuzzy controller is proposed to enhance transient stability and increase critical clearing time (CCT) in the static synchronous compensator (STATCOM). For achieving this idea, first the controller is designed based on the Lyapunov energy function. In order to avoid complexity of computation and overcome system uncertainty a neuro-fuzzy controller is proposed. In this controller, neural network determines the system rules and membership functions. In order to design a neural network and its training patterns, the energy function controller is used under various system conditions. This controller has learning abilities due to its robust fuzzy controller and neural network. Simulation results on the single-machine infinite-bus (SMIB) show that the neuro-fuzzy controller damps electromechanical oscillations and increases the critical clearing time.
  • Reza Besharati, Mozafar Bag-Mohammadi, Mashallah Abasidezfooli Page 43
    The application Layer Multicast (ALM) is an alternative and deployable approach to IP multicasting. Topology awareness link stress and delay stretch considerably, therefore it is a very important metric for ALM. This work describes a novel, highly stable and low overhead ALM approach using a binning technique to cluster nearby receivers, referred to as Bincast. Bincast uses a constant number of landmarks to cluster nearby nodes. Then, it constructs a k-ary tree between cluster members. The most stable node is selected as the head of each cluster. Cluster heads are connected to the source through a higher level tree. Detailed performance evaluation revealed that Bincast has a lower delay stretch than similar methods with approximately the same stress. Besides, it is more stable due to the selection of stable nodes as cluster heads.
  • Mansour Sheikhan, Mohsen Hatami Sadegh Page 53
    In this paper, the signal-to-noise ratio (SNR) of dynamic communication channels, using a radial basis function (RBF) neural network with time-delay structure, is estimated. The exactitude of the estimation is sufficient for systems which are based on link adaptation techniques. The proposed system for estimating SNR does not demand on having any prior knowledge about transmitted symbols. This feature of the proposed model results to save system resources. This saving is one of the benefits of proposed estimator as compared to transmitted data aided (TxDA) estimators. The performance index of the system, in terms of normalized mean squared error (NMSE) criterion, is achieved less than 0.001 for practical applications.
  • Arash Hassanpour, Sadegh Vaezzadeh Page 61
    Linear permanent magnet synchronous motors are increasingly used in industrial application. Proper performance of these motors requires appropriate design and construction of them. This paper presents the suitable topology selection and optimal design of a linear permanent magnet synchronous motor for precise applications. For this, a comprehensive analysis is done on different topologies of this kind of machines and their benefits and drawbacks are discussed in details. The proper topology is then selected and modeled via layer method. The permanent magnet poles of a typical motor are then optimized using this model to reduce the consumed permanent magnet material and therefore reducing the cost. The performance of the optimal motor is finally assessed by finite element method demonstrating the validity of the proposed method.