Elektrik - Elektronik Mühendisliği Bölümü Koleksiyonu

Permanent URI for this collectionhttps://hdl.handle.net/20.500.12469/47

Browse

Now showing 1 - 20 of 279

Citation - Scopus: 9
Vlsi Implementation of Grbf (gaussian Radial Basis Function) Networks
(IEEE, 2000) Çevikbaş, I. Can; Öğrenci, Arif Selçuk; Dündar, Günhan; Balkır, Sina; 01. Kadir Has University
A GRBF network is designed for VLSI implementation. Building blocks of the network consist mainly of analog circuits: op-amp multiplier multiplying DAC (digital to analog converter) floating resistor summer and exponentiator. Parameters of the network (center width of the Gaussian function and output layer weights) are represented digitally for convenient interfacing. It is shown that individual GRBF units allow independent tuning of center width and amplitude. Several network structures are simulated as function approximation examples and the performance is verified to be satisfactory.
Parameter quantization effects in Gaussian potential function neural networks
(World Scientific and Engineering Academy and Society, 2001) Karakuş, Erkan; Öğrenci, Arif Selçuk; Dündar, Günhan; 01. Kadir Has University
In hardware implementations of Gaussian Potential Function Neural Networks (GPFNN) deviation from ideal network parameters is inevitable because of the techniques used for parameter storage and implementation of the functions electronically resulting in loss of accuracy. This loss in accuracy can be represented by quantization of the network parameters. In order to predict this effect theoretical approaches are proposed. One-input one-output GPFNN with one hidden layer have been trained as function approximators using the Gradient Descent algorithm. After the training the network parameters (means and standard deviations of the hidden units and the connection weights) are quantized up to 16-bits in order to observe the percentage error on network output stemming from parameter quantization. Simulation results are compared with the predictions of the theoretical approach. Consequently the behaviour of the network output has been given with combined and separate parameter quantizations. Moreover given the allowed percentage error for the network a method is proposed where the minimum number of bits required for quantization of each parameter could be determined based on the theoretical predictions.
Citation - WoS: 11
Citation - Scopus: 14
Fault-Tolerant Training of Neural Networks in the Presence of Mos Transistor Mismatches
(IEEE-INST Electrical Electronics Engineers Inc, 2001) Öğrenci, Arif Selçuk; Dündar, Günhan; Balkır, Sina; 01. Kadir Has University
Analog techniques are desirable for hardware implementation of neural networks due to their numerous advantages such as small size low power and high speed. However these advantages are often offset by the difficulty in the training of analog neural network circuitry. In particular training of the circuitry by software based on hardware models is impaired by statistical variations in the integrated circuit production process resulting in performance degradation. In this paper a new paradigm of noise injection during training for the reduction of this degradation is presented. The variations at the outputs of analog neural network circuitry are modeled based on the transistor-level mismatches occurring between identically designed transistors Those variations are used as additive noise during training to increase the fault tolerance of the trained neural network. The results of this paradigm are confirmed via numerical experiments and physical measurements and are shown to be superior to the case of adding random noise during training.
Citation - Scopus: 5
A Broadband Microwave Amplifier Design by Means of Immittance Based Data Modelling Tool
(IEEE, 2002) Kilinç, Ali; Pinarbaşi, Haci; Şengül, Metin Y.; Yarman, Sıddık Binboğa; Electrical-Electronics Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
In this paper a practical broadband microwave amplifier design algorithm is introduced utilizing the immittance data-modelling tool. In the course of design first the optimum input and output terminations for the active device are produced employing the real frequency technique. Then these terminations are modelled utilizing the new immittance-modelling tool to synthesize the front-end and back-end matching networks. An example is included to exhibit the implementation of the proposed design algorithm to construct a single stage BJT amplifier over a wide frequency band. It is expected that the proposed design algorithm will find applications to realize wideband microwave amplifiers put on MMIC for mobile communication.
Citation - Scopus: 1
Edge Detection Using Steerable Filters and Cnn
(European Signal Processing Conference EUSIPCO, 2002) Özmen, Atilla; Akman, Emir Tufan; Electrical-Electronics Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
This paper proposes a new approach for edge detection using steerable filters and cellular neural networks (CNNs) where the former yields the local direction of dominant orientation and the latter provides iterative filtering. For this purpose steerable filter coefficients are used in CNN as a B template. The results are compared to the results where only CNN or steerable filters are used. As a result of this study the performance of the system can be improved since iterative filtering property of CNN and the ability of steerable filters for edge detection are used. © 2002 EUSIPCO.
Citation - WoS: 22
Citation - Scopus: 22
Selective Liquid-Liquid Extraction of Mercuric Ions by Octyl Methane Sulfonamide
(Marcel Dekker Inc, 2003) Bıçak, Niyazi; Sungur, Sana; Gazi, Mustafa; Tan, Nükhet; 01. Kadir Has University
N-octyl methane sulfonamide (OMSA) has been demonstrated to be a very efficient reagent for selective extraction of Hg(II) ions from aqueous solutions. The extraction bases on rapid reaction of OMSA with Hg(II) ions yielding mono and disulfonamido mercury compounds in ordinary conditions. Solubility of OMSA and its mercury compounds in 2-ethyl hexanol provide a clear-cut phase separation in the extraction. The solution of OMSA in 2-ethyl hexanol (0.4 mol L-1) is able to extract 82.2% of mercuric-acetate (0.4 mol L-1) in non-buffered conditions. Although the process depends on the nature of accompanying anions the distribution coefficient is reasonably high (k(d) greater than or equal to 1.27) even in the presence of chloride ions. The extraction is strictly selective and the presence of Cd(II) Zn(II) Pb(II) do not bring any interference. The extraction system works in moderate concentrations. Extracted mercury in the organic phase can be recovered by back-extraction with concentrated HCl or H2SO4 solutions. After acid treatment the organic solution of OMSA becomes regenerated without losing its activity due to reasonable hydrolytic stability of the sulfonamide linkage and it can be recycled for further extractions.
Design of Low-Pass Ladder Networks With Mixed Lumped and Distributed Elements by Means of Artificial Neural Networks
(AVES YAYINCILIK, 2003) Şengül, Metin Y.; Özmen, Atilla; Yılmaz, Melek; Electrical-Electronics Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
In this paper, calculation of parameters of low-pass ladder networks with mixed lumped and distributed elements by means of artificial neural networks is given. The results of ANN are compared with the values that are desired. It has been observed that the calculated and the desired values are extremely close to each other. So this algorith can be used to obtain the parameters that will be used to synthesize such circuits.
Yönlendirmeli Filtreler Yardımıyla Konya Bölgesi Civarındaki Gömülü Fayların Tespiti
(Doğuş Üniversitesi, 2003) Özmen, Atilla; Uçan, Osman N.; Albora, A. Muhittin; Electrical-Electronics Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
Bu makalede, yönlendirmeli filtreler jeofizik verilerin değerlendirilmesinde kullanılmışlardır. Yönlendirmeli filtreler belirli bir doğrultuda band geçiren filtrelerdir. Yönlendirmeli filtreler de, giriş görüntüsündeki farklı yönlerdeki kenarların elde edilmesi için, görüntü ilk önce farklı yönlere sahip temel filtrelerden geçirilir ve daha sonra yönelim alt bandlarına ayrılır. Bu çalışmada, yönlendirmeli filtrelerin başarımını görebilmek için, çeşitli açılara sahip sentetik datalar ele alınmış ve kenar belirlemesi yapılmıştır. Arazi çalışması olarak, Konya bölgesinin gravite anomali haritasını kullandık. Gömülü durumda bulunan fayların oluşturduğu anomaliler farklı yönler için incelenmiş ve bölgenin oluşturulan fay haritası jeolojik bilgilerle karşılaştırılmıştır.
Sliding Mode Controller Solution for the Shallow Submerged Operation Ok a Submarine
(IFAC Secretariat, 2003) Demirci, Ufuk; Kerestecioğlu, Feza; Computer Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
In this paper a submarine controller is presented which can accommodate the sea wave effects on a submarine a(shallow water operation. Sliding mode method is implemented in a way that the robustness of the controller increased with respect to disturbance distribution vector in order to perform the depth control of a shallow submerged submarine under sea wave disturbances. Designed controller kept the submarine performance within acceptable limits. Copyright © 2003 IFAC.
Citation - Scopus: 4
Pilot-Aided Bayesian Mmse Channel Estimation for Ofdm Systems: Algorithm and Performance Analysis
(2004) Şenol, Habib; Cirpan, Hakan Ali; Panayırcı, Erdal; Computer Engineering; Electrical-Electronics Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
This paper proposes a computationally efficient pilot-aided minimum mean square error (MMSE) channel estimation algorithm for OFDM systems. The proposed approach employs a convenient representation of the discrete multipath fading channel based on the Karhunen-Loeve (KL) orthogonal expansion and estimates uncorrelated series expansion coefficients. Moreover optimal rank reduction is achieved in the proposed approach by exploiting the optimal truncation property of the KL expansion resulting in a smaller computational load on the estimation algorithm. The performance of the proposed approach is studied through analytical and experimental results. We first consider the stochastic Cramer-Rao bound and derive the closed-form expression for the random KL coefficients. We then exploit the performance of the MMSE channel estimator based on the evaluation of minimum Bayesian MSE. © 2004 IEEE.
Citation - Scopus: 1
An Open Software Architecture of Neural Networks: Neurosoft
(2004) Öğrenci, Arif Selçuk; Arsan, Taner; Saydam, Tuncay; Computer Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
Software architecture of generic distributed neural networks and its relevant information model have been developed. Principles of on-line architecture building training controlling (managing) and topological optimization guidelines are provided and extensively discussed.
Citation - Scopus: 7
Fault Tolerant Control With Re-Configuring Sliding-Mode Schemes
(2005) Demirci, Ufuk; Kerestecioğlu, Feza; Computer Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
In this paper a controller design method for linear MIMO systems is presented which a sliding mode controller is reconfigured in case of system faults. Faults are detected with the residual vector generated from a standard linear observer. Once a fault has been detected the fault distribution matrix can be obtained and used to update the corrective or equivalent control parts of the sliding mode controller. As a result fault tolerant adaptive controllers keep the system performance within acceptable limits or at least avoids the system to wind-up. © TÜBITAK.
Fault Estimation Of Trakya And Marmara Sea Regions Using 2d Gabor Filtering [2 Boyutlu Gabor Filtre Yöntemi Uygulayarak Trakya ve Marmara Denizindeki Fay Hatlarının Saptanması]
(2005) Özmen, Atilla; Erdoğan, Didem; Uçan, Osman Nuri; Albora, Ali Muhittin; Electrical-Electronics Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
In this paper we have applied 2D Gabor filtering to gravity and magnetic anomalies in estimation of discontinuities. Gabor filtering is an effective separation method compared to others having steerable and frequency parameter properties. We have found new faults using Gabor filtering for gravity and magnetic anomalies of Marmara Sea. © 2005 IEEE.
Citation - WoS: 8
Citation - Scopus: 11
Signature verification using conic section function neural network
(Springer-Verlag Berlin, 2005) Şenol, Canan; Yıldırım, Tülay; 01. Kadir Has University
This paper presents a new approach for off-line signature verification based on a hybrid neural network (Conic Section Function Neural Network-CSFNN). Artificial Neural Networks (ANNs) have recently become a very important method for classification and verification problems. In this work CSFNN was proposed for the signature verification and compared with two well known neural network architectures (Multilayer Perceptron-MLP and Radial Basis Function-RBF Networks). The proposed system was trained and tested on a signature database consisting of a total of 304 signature images taken from 8 different persons. A total of 256 samples (32 samples for each person) for training and 48 fake samples (6 fake samples belonging to each person) for testing were used. The results were presented and the comparisons were also made in terms of FAR (False Acceptance Rate) and FRR (False Rejection Rate).
Citation - Scopus: 13
A Low-Complexity Kl Expansion-Based Channel Estimator for Ofdm Systems
(2005) Şenol, Habib; Cirpan, Hakan Ali; Panayırcı, Erdal; Computer Engineering; Electrical-Electronics Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
This paper first proposes a computationally efficient pilot-aided linear minimum mean square error (MMSE) batch channel estimation algorithm for OFDM systems in unknown wireless fading channels. The proposed approach employs a convenient representation of the discrete multipath fading channel based on the Karhunen-Loeve (KL) orthogonal expansion and finds MMSE estimates of the uncorrelated KL series expansion coefficients. Based on such an expansion no matrix inversion is required in the proposed MMSE estimator. Moreover optimal rank reduction is achieved by exploiting the optimal truncation property of the KL expansion resulting in a smaller computational load on the estimation algorithm. The performance of the proposed approach is studied through analytical and experimental results. We then consider the stochastic Cramér-Rao bound and derive the closed-form expression for the random KL coefficients and consequently exploit the performance of the MMSE channel estimator based on the evaluation of minimum Bayesian MSE. We also analyze the effect of a modelling mismatch on the estimator performance. To further reduce the complexity we extend the batch linear MMSE to the sequential linear MMSE estimator. With the fast convergence property and the simple structure the sequential linear MMSE estimator provides an attractive alternative to the implementation of channel estimator.
Citation - WoS: 1
Support Vector Machines Based Target Tracking Techniques
(IEEE, 2006) Özer, Sedat; Çırpan, Hakan Ali; Kabaoğlu, Nihat; 01. Kadir Has University
This paper addresses the problem of aplying powerful statistical pattern classification algorithms based on kernels to target tracking. Rather than directly adapting a recognizer we develop a localizer directly using the regression form of the Support Vector Machines (SVM). The proposed approach considers using dynamic model together as feature vectors and makes the hyperplane and the support vectors follow the changes in these features. The performance of the tracker is demostrated in a sensor network scenario with a moving target in a polynomial route.
Citation - Scopus: 2
Support Vector Machines Based Target Tracking Techniques [destek Vektör Makineleri Tabanlı Hedef Takip Yöntemleri]
(2006) Özer, Sedat; Cirpan, Hakan Ali; Kabaoğlu, Nihat; 01. Kadir Has University
This paper addresses the problem of aplying powerful statistical pattern classification algorithms based on kernels to target tracking. Rather than directly adapting a recognizer we develop a localizer directly using the regression form of the Support Vector Machines (SVM). The proposed approach considers using dynamic model together as feature vectors and makes the hyperplane and the support vectors follow the changes in these features. The performance of the tracker is demostrated in a sensor network scenario with a moving target in a polynomial route. © 2006 IEEE.
A Numerical Method for Frequency Determination in the Astable Cellular Neural Networks With Opposite-Sign Templates
(IEEE, 2006) Özmen, Atilla; Tander, Baran; Computer Engineering; Electrical-Electronics Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
In this study a numerical method is proposed to determine the oscillation frequencies in the astable cellular neural networks with opposite-sign templates [1]. This method depends on the training of a multilayer perceptron that uses various template coefficients and the correspondant frequency values as inputs and outputs. First of all a frequency surface is obtained from templates and then training samples are picked from this surface in order to apply to multilayer perceptron. The effects of the template coefficients to the oscillation frequencies are also investigated. Furthermore an oscillator design is carried out for simulation and the performance as well as the advantages of the proposed method are evaluated.
Citation - WoS: 2
Citation - Scopus: 2
Linear Expansions for Frequency Selective Channels in Ofdm
(Elsevier GMBH Urban & Fischer Verlag, 2006) Şenol, Hande; Çırpan, Hakan Ali; Panayırcı, Erdal; Electrical-Electronics Engineering; 05. Faculty of Engineering and Natural Sciences; 01. Kadir Has University
Modeling the frequency selective fading channels as random processes we employ a linear expansion based on the Karhumen-Loeve (KL) series representation involving a complete set of orthogonal deterministic vectors with a corresponding uncorrelated random coefficients. Focusing on OFDM transmissions through frequency selective fading this paper pursues a computationally efficient pilot-aided linear minimum mean square error (MMSE) uncorrelated KL series expansion coefficients estimation algorithm. Based on such an expansion no matrix inversion is required in the proposed MMSE estimator. Moreover truncation in the linear expansion of channel is achieved by exploiting the optimal truncation property of the KL expansion resulting in a smaller computational load on the estimation algorithm. The performance of the proposed approach is studied through analytical and experimental results. We first exploit the performance of the MMSE channel estimator based on the evaluation of minimum Bayesian MSE. We also provide performance analysis results studying the influence of the effect of SNR and correlation mismatch on the estimator performance. Simulation results confirm our theoretical results and illustrate that the proposed algorithm is capable of tracking fast fading and improving performance. (c) 2005 Elsevier GmbH. All rights reserved.
Near Field Parameter Estimation Of Moving Sources With Recursive Expectation Maximization Algorithm [yinelemeli Beklenti/en Büyükleme Algoritması ile Hareketli Kaynakların Yakın-alan Parametrelerinin Kestirimi]
(2006) Çekli, Serap; Çekli, Erdinç; Kabaoğlu, Nihat; Cirpan, Hakan Ali; 01. Kadir Has University
In this paper maximum likelihood (ML) estimator is proposed for the joint estimation of the direction of arrival (DOA) and range parameters of moving sources in the near-field of the antenna array. ML estimation algorithm is presented for deterministic signal model. Recursive form of the expectation maximization (REM) algoritm is suggested for the estimation of the near-field parameters because there is not closed form solutions for the maximum likelihood functions. Moreover simulation results of the suggested algorithm are presented. © 2006 IEEE.

Browse

Browsing Elektrik - Elektronik Mühendisliği Bölümü Koleksiyonu by Issue Date