مجله مهندسی برق و الکترونیک ایران
سال سیزدهم شماره 2 (تابستان 1395)

In this paper¡ the design and simulation of an antenna for SAR applications¡ with specified imaging characteristics¡ is proposed. The antenna should represent a wideband impedance bandwidth and also a high gain; so the design of a microstrip array antenna with proximity coupling feed and U-shaped slot patch is considered. First¡ the main design parameters of such an antenna element are recognized by parametric study. Then the antenna element¡ with more than 16% impedance bandwidth¡ is designed¡ and the simulation results are presented and compared with measurement results. Finally¡ a preliminary design for the planar array is theoretically proposed regarding to the dimensions and the current distribution¡ which complies with the desired antenna characteristics.

In this paper¡ the spatial MIMO channel modeling (SCM) is introduced and the required parameters for this modeling are studied. Then¡ the MIMO channel gains and capacity are analyzed. The transmitter and receiver antenna patterns are also analyzed. Next¡ we are focused on the effects of the transmitter and receiver antenna arrays separations¡ the arrays length¡ and the angles spread on the antenna patterns and channel capacity. The simulation results of the SCM show that the MIMO channel capacity increases as the transmitter array separation increases. However¡ it cannot be improved as the receiver array separation increases.

The challenging characteristics of Smart Grid such as the vast size of covered area and tight requirements of reliability (higher than 0.98) and delay (in terms of second and millisecond) impede the procedure of communication network designing. This paper investigates the problem of network designing subject to the reliability constraint. The main contribution is dividing the problem into two sub-problem¡ Primary Topology Design Sub-problem (PTDS) and Reliability Assurance Sub-problem (RAS). In PTDS a primary network of routers is devised to cover all network nodes using linear programming and RAS makes the primary topology reliable by adding more routers through a repetitive procedure. The main idea of RAS is balancing degrees of nodes in the network graph. Results show that the provided approach devises a fairly optimal network in an acceptable time.

Transmission opportunity limit and contention window are two influenced parameters in accessing the media in the networks based IEEE 802.11e standard. In these networks¡ every wireless station by setting of its contention window value¡ determines the chance of access to wireless media. Furthermore¡ a transmission opportunity limit provides a period of transmission without contact for a station. In order to increase the productivity of the wireless media¡ these two parameters can be determined by considering the network variable conditions in the dynamic form. In this regard¡ since Games Theory is a robust tool for modeling¡ analysis and optimization of competitive environment with shared resources¡ in this research¡ the problem of simultaneous setting of transmission opportunity and contention window in the framework of Game theory have been surveyed and this problem has been modeled in the format of a game. In this Game¡ by defining the consequence function¡ every wireless station plays a role of a player and selects its strategy in setting these two parameters by maximizing its consequence function and in accordance with the quality of services requirements and increasing the productivity of wireless media. Simulation results show that the proposed method has caused to improve access to the channel and relative fairness retention.

In this paper we propose the spatial sparsity based WCSSDOA method for multi speakers'' Direction of arrival estimation. In the proposed method the sparse modeling is done based on the sensor signals'' correlation matrix¡ which leads to low computational complexity. In this method the SVD decomposition of the noise covariance matrix is proposed to reach the free noise sparse model¡ which leads to high accurate method in the low SNR with respect to other methods. Simulation results show that contrary to MUSIC¡ Minimum-norm¡ and SSDOA methods¡ the proposed method could detect the direction of two narrowband sources with distance in -5dB SNR¡ six narrowband sources with distance in -15dB SNR¡ and two speakers with distance in 10dB SNR. Also¡ the computational complexity of the proposed method is reduced 45-85% with respect to SSDOA method.

The selection of a suitable feature is an important and indispensible issue in real-time surveillance systems employing a network of cameras to achieve rapid and accurate results. We have used the concept of color image histograms¡ as a simple but efficient method¡ to track people in different cameras in this research. Employing the image histogram for tracking may not be robust as it is sensitive to the environmental brightness changes and the human body shape and size. In this paper¡ we propose a novel method employing the Cumulative Brightness Function (CBF) to attenuate the effects of brightness changes in tracking objects moving in the camera networks. In the proposed method a fuzzy system is used to partition the field of view of each camera to cope with the size variations of people in different cameras. Then the body of each subject is divided into three portions¡ namely the head¡ the torso and the bottom section¡ with the prior knowledge about the human physiology¡ to have a better distinction between variance people appearance color. Finally¡ the histograms of the body sections with respect to the distance of subject from each camera are used for subject identification¡ reidentification and tracking. The results achieved by the proposed method in various environmental conditions show its robustness and effectiveness for tracking people in a network of cameras.

In this paper a destriping technique is proposed for pushbroom-type satellite imaging systems. This technique is based on Moment Matching algorithm and assumes a linear response for each detector of imaging system. In the most of the works in this field¡ the offset parameter in detectors’ response is neglected and images are corrected using only an estimation of gain parameter. Proposed method considers an estimated offset for detectors'' response and corrects image. Simulations results show that destriping with proposed method is much better than the gain method (adaptive moment matching method). In addition¡ comparing the edge density of the results and a perfect reference image shows superiority of the proposed algorithm.

Image segmentation is one of the most important and difficult steps in machine vision problems and achieving the desired results often requires satisfaction of different objectives. One approach to face this situation uses multi-objective fuzzy clustering of pixels in the feature space. This paper proposes a new strategy for search within the family of multiobjective differential evolution algorithms with the purpose of finding optimal partitions of pixels. Based on this¡ all of the encoded clusters in the population of one generation are clustered and each centroid of one donor vector is made with centers from a unique cluster. Through the search space dimension is reduced and searching for each centroid focused on the different area of input space while cluster centers of one fuzzy partition preserve separation. The performance of the proposed method is compared with two other multi-objective fuzzy clustering methods to segment a number of images from the Berkeley segmentation database. Visual and quantitative evaluations show that the proposed method has a better match with the ground truths than the other methods.

One of the most preprocessing steps before the classification of hyperspectral images is supervised feature extraction. Because obtaining the training samples is hard and time consuming¡ the number of available training samples is limited. We propose a supervised feature extraction method in this paper that is efficient in small sample size situation. The proposed method¡ which is called weighted feature line embedding (WFLE)¡ uses the feature line concepts for production of virtual training samples and then¡ uses them for estimation of within-class and between-class scatter matrices. The new idea of WFLE is based on more correction on the non-appropriate and abnormal samples through weighting process in estimation of scatter matrices. The WFLE is compared with some popular and state-of-the-art feature extraction methods such as LDA¡ GDA¡ NWFE¡ NPE¡ LPP and NFLE. The experimental results show the good performance of WFLE in comparison with other methods in small sample size situation.

Reducing the locking time or settling time is one of the major challenges in the design of Delay Locked Loop (DLL) based frequency synthesizer. In this paper a common structure for DLL based frequency synthesizer is considered in which the number of delay cells in the direct path is specified. Then¡ the designed delay locked loop is optimized using genetic algorithm (GA). GA changes the phase-voltage gain coefficients of the delay cells for achieving the best locking (settling) time. Typical DLL with a reference frequency of 100 MHz and 8 delay cells is studied. Simulation results is shown the proposed structure is locked in 0.58 mu.

A nonlinear model predictive control (NMPC) algorithm based on neural network is designed for boiler- turbine system. The boiler–turbine system presents a challenging control problem owing to its severe nonlinearity over a wide operation range¡ tight operating constraints on control move and strong coupling among variables. The nonlinear system is identified by MLP neural network and neural predictive control is designed by MIMO neural network model. Using neural network as predictive model in predictive control results in steady state error. A disturbance model is used in neural predictive model for elimination of steady state error and disturbance rejection. Simulation results on a boiler turbine system illustrate that a satisfactory closed-loop performance and offset-free property can be achieved by using the proposed method.

Due to volatility of spot power prices and in order to manage risk of a Generating Company (GenCo)¡ this paper addresses determination of optimal quantity of bilateral forward contracts which can be formulated as an optimization problem. In this framework¡ in addition to selling electricity to spot market¡ bilateral forward contracts can be traded between Gencos and customers. Finding an optimal trade-off between quantity of selling power to the forward market and the spot market is an important decision making problem. In this paper¡ considering spot price uncertainty and according to the definite forward contract price an analytical method is utilized to solve this stochastic optimization problem in the Pay-as-Bid auction. The Lagrange multiplier method is used to obtain numerical solutions for optimal amounts of energy which should be sold on the forward market and spot market. Finally¡ the effect of uncertainty in the spot market price and cost function coefficients of the studied GenCo on the optimization problem results have been investigated using a sensitivity analysis.

Most of the recent studies have tried to create diversity in primary results and then applied a consensus function over all the obtained results to combine the weak partitions. In this paper a clustering ensemble method is proposed which is based on a subset of primary clusters. The main idea behind this method is using more stable clusters in the ensemble. The stability is applied as a goodness measure of the clusters. The clusters which satisfy a threshold of this measure are selected to participate in the ensemble. For combining the chosen clusters¡ a co-association based consensus function is applied. A new EAC based method which is called Extended Evidence Accumulation Clustering¡ EEAC¡ is proposed for constructing the Co-association Matrix from the subset of clusters. The proposed method is evaluated on five different UCI repository data sets. The empirical studies show the significant improvement of the proposed method in comparison with other ones.