Elektronik ve Haberleşme Mühendisliği Bölümü Tezleri

Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/179

Browse

Now showing 1 - 20 of 115

Simulation of free space optical links under different atmospheric conditions
(2011) Öztürk, Korhan
This thesis contains a feasibility study for the free space optical links, which would be installed in Turkey. In first chapters, some basic information and the background theory of the atmospheric channel are given. Predefined optical links are simulated with PCModWin-MODTRAN radiative transfer code to investigate the link availability under the influence of different weather conditions. In addition, numerical calculations according to some experimentally approved models are also performed and these results are compared with the results of simulations. For all these simulations and calculations, big amount of meteorological data was collected from the airport stations, which are in different cities in Turkey. The parameters used in calculations were supplied from the commercially available FSO systems. In conclusion, the effectiveness of FSO links in Turkey is discussed.
Comparison of consensus algorithms in wireless sensor networks
(2019) Sulaiman, Abdulmajeed
Kablosuz sensör ağları geniş bir uygulama yelpazesinde, çok farklı türden veriyi takip etmek için kullanılmaktadır. Yakın gelecekte, Nesnelerin Interneti kavramının geniş kabulü ile kullanım alanı daha da artacak olan bu ağlarda önemli problemlerden biri de ağdaki sensörlerin takip edilen değerler konusunda hemfikir olmalarını sağlamaktır. Bu amaca yönelik olarak farklı uzlaşı algoritmaları geliştirilmiştir. Bu algoritmalar, düşük hesaplama karmaşıklığı, hızlı yakınsama, düşük enerji sarfiyatı gibi farklı kriterleri sağlamayı amaçlamaktadırlar. Bunun yanı sıra, hareketli otonom cihazların belli bir noktada buluşmaları gibi başka alanlarda da uzlaşı algoritmaları kullanılmaktadır. Bu tezde amacımız, farklı kriterlere göre tasarlanmış uzlaşı algoritmalarının benzetimler ışığında karşılaştırmalarını sunmak, ve kullanım alanlarına göre her birinin diğerlerine göre avantajlarını ve dezavantajlarını ortaya çıkarmaktır.
Power scintillation for super lorentz gaussian beam evaluation using the random phase screen method
(2018) Al-Grairi, Hussein Thary Khamees
Bu tezde, Süper Lorentz Gauss (SLG) ışık hüzmelerinin belirli bir alıcı açıklık yarıçapı için ortalama optik güç pırıldama indeksi incelenmiştir. Orta dereceli türbülanslı ve zayıf dereceli türbülanslı atmosferde SLG ışık hüzmesinin davranışı rastgele fazlı pencere yöntemi kullanılarak nümerik olarak analiz edilmiştir. Ortalama güç pırıldama indeksi için kaynak boyutu ve dalga boyu değişimi gözlemlenmiştir. Buna ilaveten alıcı düzleminin merkez noktasında pırıldama indeksi ile güç pırıldama indeksi dalga boyu değişimlerine göre karşılaştırılmıştır. Nokta-pırıldama indeksi, daha küçük dalga boyu değerleri ile azalmasına rağmen, güç pırıldama indeksi farklı dalga boylarında hemen hemen aynıdır. Sayısal sonuçlara göre, güç pırıldama indeksinin de nokta-pırıldama indeksi gibi, artan kaynak boyutları ile azaldığı gözlemlenmiştir. Tüm yayılma parametrelerine bağlı olarak alınan sonuçlara göre, SLG ışık hüzmesinin güç pırıldama indeksi, SLG ışık hüzmesinin nokta-pırıldama indeksinden daha küçüktür. Ayrıca, alıcı düzleminde ortalama ışık şiddeti yoğunluğu profilleri, farklı yayılım mesafesi için çizdirilmiştir. Alınan sonuçlara göre artan mesafelerde ışık şiddetinin azaldığı ve genişlediği gözlemlenmiştir. Alıcı düzleminde, SLG ışık hüzmesinin ortalama ışık şiddetinin iki ve üç boyutlu profilleri ayrıntılı olarak incelenmiştir. Alıcıdaki ışık şiddeti, farklı kaynak boyutlarında ve dalga boylarında incelenmiştir ve alıcıdaki ışık şiddetinin artan yayılma mesafesinde azaltıldığı gözlemlenmiştir. Bu çalışmada, alıcıdaki pırıldama indeksi ve ışık şiddetinin hesaplanması için MATLAB kodları kullanılmıştır. Ayrıca, farklı türbülans değerlerinde, SLG ışık hüzmesi derecesi değiştirilerek, Gauss ışık hüzmesi için pırıldama indeksi değerleri elde edilerek, alınan sonuçların doğruluğu, Gauss ışık hüzmesi için literatürde daha önce alınan sonuçlar ile karşılaştırılarak test edilmiştir.
BER and SNR performance analysis of 2-ASK & 4-ASK modulated truncated bessel beams on different receiver apertures in FSO communications links
(Çankaya Üniversitesi, 2017) Ağar, Tevfik
In this thesis, varying turbulent atmosphere environment are simulated random phase screen method. 2-ASK & 4-ASK modulated truncated Bessel beam are generated and propagated through turbulent atmosphere utilizing random phase screens. In MATLAB environment error counting method is applied then simulation results are obtained corresponding to BER and SNR. On the other hand, the contributions of enlargement receiver aperture on BER is examined. Also, to comprehend the free space optical (FSO) communication architecture in detail and to support the future experimental studies, FSO communication transmitter and receiver systems are designed.
Analysis and observation on quality of service for data networks of turkish telecommunication infrastructure
(2008) Demirel, Mustafa
In this thesis, Jitter, Delay and Packet Loss values are measured for different points on TT IP/MPLS Backbone and compared with acceptable values. Measurements are made by using TT’s Alcatel 7750-SR12 devices which run in Ankara, Konya, Kahramanmaraş and Van. During measurements the device running in Ankara is always used as resource.Each of the other points are used as targets.The transmission distances between the devices measurements are as Ankara-Dikmen distance 10 Km and Ankara-Konya 300 Km, Ankara-Kahramanmaraş 602 Km and finally Ankara-Van 1248 Km. In this way, the connection between Jitter, Packet Loss and Delay Values and transmission distance is determined During tests every measurement is performed using 1500 and 4000 Bytes divided packets.The measurement values are taken using ping command on a software called SMART which can reach all devices on TT IP/MPLS Backbone. Measurements made are calculated in a day time zone. The connection of the software by which we perform measurements to the backbone is 100 Mbps. In the measurements taken, increase in Jitter and Delay value is observed when there is an increase in packet size. At the same time, increase in Jitter and Delay value is observed when there is an increase in transmisson distance. The packet loss value is evaluated as %0 in all of the tests made on TT IP/MPLS Backbone.
Simulative and real time DC motor speed control
(2016) Mahmood, Ahmed Imad
DC motor velocity control is one of the most commonly encountered control applications in literature as DC machines are exclusively used in industrial applications. The DC motors are assumed to have linear and stable characteristics, and for this reason they are supposed to be controlled easily. However that is not the case in real time applications. This thesis provides a real time controller design approach in cooperation with system identification. Primarily the linear mathematical model of an experimental DC motor speed control system set-up is constructed in simulation and this model is used to obtain a controllerfor the speed control mission yet the controller obtained in the simulation turn out to be unsuccessful in the real time application. Hence, a real time system identification step is executed to attain an approximate system linear model and it is observed that identified system structure varied from the mathematical model significantly. Finally, A new controller is designed for the identified system and it verified by different tests that this controller operates comparably well in real time applications.
General optical beam formulation and their propagation in atmospheric optical telecommunication links
(2007) Yazıcıoğlu, Canan
In this thesis, we have developed a general source beam to combine many types of different beams such as Bessel, Bessel Gaussian, Laguerre, Laguerre Gaussian, Ince Gaussian, dark hollow, bottle, super Gaussian, Lorentz, flat-topped, Hermite-sinusoidal- Gaussian, sinusoidal-Gaussian like cos-Gaussian, sine-Gaussian, cosh-Gaussian, sinh- Gaussian, annular and their higher order modes with their truncated, elegant and elliptical versions in a single expression. Also we developed a Matlab code to plot intensity distributions at source and receiver. Using this Matlab code and the general source beam formulation, all the mentioned beams? source equations are obtained, compared with their existing forms in the literature and intensity patterns of some of them are plotted at source plane. Using a part of this formulation and the Matlab simulation, intensity is calculated and plotted intensity distributions of Hermite-sinusoidal-Gaussian, sinusoidal- Gaussian, annular and higher-order annular beams at different propagation distances in atmospheric turbulence
Radius of curvature of laguerre-gaussian beam
(2012) Çay, Saim
In this thesis, the radius of curvature of Laguerre-Gaussian beam is formulated for turbulent atmosphere and analyzed numerically for various radial and angular mode numbers, the source size, propagation distance and wavelength in moderate, high and no turbulence levels. Results have shown that Laguerre-Gaussian beam approximates to Gaussian beam and radius of curvature of Laguerre-Gaussian beam reduces with increasing turbulence levels. The results have also shown that the radius of curvature of Laguerre-Gaussian beam increases with increasing source sizes and changes slowly with longer wavelengths.
Recursive two-stage evolutionary programming based on similarity measures
(2013) Önal, Turgut
This thesis is devoted to deeper research about the characteristics of the Metaheuristic search algorithm “Recursive Two-Stage Evolutionary Programming” (RTEP). This algorithm is a powerful Metaheuristic where only mutation operation is used in a specific style in order to produce better candidate solutions for the dealt optimization problem. The algorithm has many different parameters and procedures that can be adapted in order to obtain better solutions for the optimization problems. Especially, the distance measure between the chromosomes, how the selection operation will be carried out and the style of the mutation operation are important subjects of this search algorithm. In this thesis, the contribution of parameter and procedure changes for these subjects in the efficiency of the algorithm is investigated.
Trajectory generation for open/close gap maneuvers in vehicle strings
(Çankaya Üniversitesi, 2017) Çankaya, Erkam
Cooperative adaptive cruise control (CACC) is an advanced technology allowing vehicle following at a small inter-vehicle spacing. In its classical usage, CACC assumes that vehicles are arranged in the form of a vehicle string and follow each other at a velocity-dependent distance. Nonetheless, practical driving situations include the case of lane changes, where vehicles can join or leave a vehicle string. In such case, it is required that gaps for joining vehicles are provided or gaps after leaving vehicles are closed in order to ensure safe and efficient driving. This thesis is concerned with gap opening and closing maneuvers in vehicle strings. Introducing a suitable control architecture, gap opening and closing maneuvers can be realized by the generation of feedforward input signals. To this end, the first contribution of the thesis is the development of five methods for the computation and representation of gap opening and closing trajectories that fulfill additional safety and comfort constraints. The first method is based on the solution of an optimal control problem, the second method uses a polynomial trajectory and plant inversion, the third method concatenates three polynomials and uses nonlinear programming to determine the polynomial coefficients, the fourth method uses a high-order polynomial and the fifth method uses concatenated polynomials in order to approximate the optimal control solution. A simulation study shows that the fifth method is particularly useful in practical applications since it computes trajectories that approximate the optimal control solution in real-time. The second contribution of the thesis is the implementation of a vehicle model that realizes CACC and additional feedforward signals in the form of a Matlab S-function.
Ship detection and classification using type of convolution neural networks
(2020) Kunt, Yusuf
Gemi sınıflandırma ve tanıma sistemleri, denize kıyısı ve boğaz geçişleri yoğun olan ülkeler de ulusal ve yerel savunma, gemi trafik kontrolü, kaçak balıkçılık, korsan, insan tacirleri ve küresel ticaret zinciri gibi sorun oluşturabilecek birçok alanda kullanılmaktadır. Bu sorunlara erken müdahale edebilmek için gemi sınıflandırma ve tanıma ile ilgili farklı farklı çözüm metodları geliştirilmiştir. Bu metodlar üzerinde yapılan araştırmalarda yoğun olarak iki gruba ayrılmıştır. Bunlar Uydu ve SAR görüntüleri üzerine yapılan çalışmalardır. Bu kaynaklar üzerinde yapılan çalışma sayısı fazla olsada gelen görüntüler herkesin kullanımına açık değildir. Ayrıca çözünürlükleri gemi sınıflandırma ve tanıma için yetersiz kalmaktadır. Bu metodlara alternatif çözüm olarak derin öğrenme teknolojisi olan CNN' ler oldukça popüler hale gelmiştir. Geçtiğimiz 30 yıl boyunca, bilgisayar görme teknolojileri görsel görevlerde insanlara yardım etmekte zorlandı. Ancak, bugün derin öğrenme teknolojisindeki büyük ilerlemeler, bilgisayarların görüntüleri bir kişi kadar başarılı ve hatta daha iyi işleme yeteneğini sağlamıştır. Tezimizin CNN mimarileri üzerine kurulmasının sebeplerinden birisi de CNN modellerin sınıflandırma ve tanıma problemlerindeki senaryolara göre kolaylıkla sınıf sayılarının genişletilebilir olmasıdır. Bu tezde gemi sınıflandırma için VGG16, VGG 19, DenseNet121, Xception gibi literatürdeki en popüler derin öğrenme CNN mimarileri kullanılmıştır. Ayrıca gemi tanıma için ise YOLOv3 gibi yüksek performans elde edilen CNN mimari tercih edilmiştir. Tezimizde bu modellerin sonuçları karşılaştırılmıştır. Eğitim ve test için kendimizin oluşturduğu 6 adet gemi den oluşan veri setimiz kullanılmıştır. CNN modelleri çok fazla sayıda eğitim seti gerektirdiği için öğrenme aktarım teknolojisi, veri yetersizliği sorununu ortadan kaldırmak amacıyla bu modellerde uygulanmıştır. Tezimizin literatüre olan katkısını, gemi sınıflandırma problemleri için CNN model mimarilerin karşılaştırılması, CNN model mimarisindeki katman sayılarının etkilerinin araştırılması ve bu probleme yönelik uygun hiper parametre seçimi olarak sıralayabiliriz. Gemi tanıma için ise etiketli veri sayısının performans üzerindeki etki olarak değerlendirebiliriz.
Improvements in the finite element method and method of moments for the solution of electromagnetic problems
(Çankaya Üniversitesi, 2017) Albayaty, Abdullah Omran Ali
In this thesis, two numerical techniques, the finite element method (FEM) and the method of moments (MoM) are improved for solving electromagnetic boundary value problems governed by differential and integral equations. The FEM is a numerical technique used to solve differential equations. The solution domain is divided into small intervals (elements), and the unknown function is calculated by finding an approximate solution by a truncated series of different shape functions. This approximate solution can be replaced with the original function in the differential equation. As a result, we get the matrix of equations that can be solved to obtain separate solution to the problem. In this thesis, we used the sigmoid function as a novel application, and compare the accuracy of FEM with the sigmoid function against the well-known linear and the step functions. Two examples of electromagnetic problems are presented. It is concluded that the sigmoid function under specific conditions yields the most accurate results. For MoM, this thesis presents a new mathematical algorithm for the solution of electromagnetic problems. This algorithm is schemed to be suitable for solving the singularity that exists in the solution matrix. This adaptive integration algorithm aims to avoid the singularity in the evaluation of the integral so called the Cauchy Principal Value integral. An example electrostatic problem is presented and it is proved that the novel method is accurate.
Development of air data computation function of a combined air data and aoa computera
(2011) Ilğın, Rafet
In this thesis, Air Data Computer part of a Combined Air Data System (CADS) and the simulator environment to test the developed CADS are developed on standard personal computers. Normally, a CADS system on an aircraft is composed of two separate equipments, the Air Data Computer (ADC) and the Angle of Attack (AOA) system. Therefore the developed CADS system combines both functionalities in an integral manner on a card. This approach not only reduces the volume but the total cost of the CADS system as well.
Fault-tolerant supervisory control of discrete event systems: Methods and examples
(2018) Acar, Ayşe Nur
Hatalar kontrol edilemeyen olaylardır. Sistemde bir anda meydana gelirler ve sistemin çalışma düzenin negatif yönde etkilerler. Genellikle, ayrık olaylı sistemlerde (DES) hata olması durumunda, ilgili olayların veya operasyonların bir daha oluşmama ihtimali vardır. Bu tezde hata içeren ayrık olaylı sistemlerin denetleyici kontrolü çalışılmıştır. Bu vesile ile iki ana konsept incelenmiştir: Hataya dayanıklılık ve hata kurtarıcı kontrol. Hataya dayanıklı kontrole ilişkin, sistemin hem normal durumunda hem de hatalı durumda tek bir kontrolör kullanılması amaçlanmıştır. Buradan yola çıkarak, önce hem hatalı hem de hatasız durumda verilen tavrı çalıştırabilen bir spesifikasyonun gerekli ve uygun koşulları tanımlanmıştır. Daha sonra gösterilmiştir ki, tanımlanan koşulların ihlal edildiği durumlarda supremal bir arızaya dayanıklı alt dil belirlenmesi mümkündür. Son olarak bu alt dilin bulunabilmesini sağlayan bir algoritma geliştirilmiş ve bunun doğruluğu kanıtlanmıştır. Var olan diğer çalışmalardan farklı olarak, bizim geliştirdiğimiz hataya dayanıklı kontrolör, hatanın oluşmasına ve sistemin herhangi bir zamanda tamirine izin vermektedir. Hata kurtarıcı kontrole ilişkin, ayrık olaylı sistemlerin operasyonunda hem hata sonrası durumuna hem de tamir sonrası durumunun kontrolü için çalışılmıştır. Öncelikle DES'e yönelik arıza kurtarma için yeni bir model geliştirilmiştir. Özellikle, bir arızadan sonra bozuk bir spesifikasyona göre işlemine devam eden ve sonunda da arızadan sonra istenen bir davranışa yönelen bir arıza oluşumuna kadar, belirtilen nominal sistem davranışını takip eden bir arıza kurtarma denetçisi hesapladık. Ardından, yöntemimizin sistem tamirine de uygulanabileceği gösterilmiş ve isteğe bağlı sayıda arıza oluşumu ve sistem tamiri durumunu kontrol edebilebilen bir denetleyiçi oluşturulmasını sağlayan yinelemeli bir algoritma önerilmiştir. Son olarak, arıza kurtarma ve tamir yöntemi, birçok ve farklı arıza ve tamirler ile birlikte sunulmuştur. Sonuç olarak, arızasız durumlarda belirtilen nominal sistem davranışını takip eden, her bir arıza tipi için istenen bozuk davranışa yönelen ve ilgili tamir işleminden sonra nominal davranışı kurtaran bir denetçi elde edilmiştir. Geliştirilen tüm yöntemler ayrı ayrı oluşturduğumuz bir küçük üretim sistemi örneğiyle gösterilmiş ve tüm sonuçlar aynı örnek üzerinde açıklanmıştır.
Driver design for asynchronous motors: Digital signal processing, control and data acquisition
(2012) Öz, Habib Özer
In this research, a three-phase AC induction motor driver training set is designed for students and educators. The AC induction motor driver training set design consists of two main parts: the power electronic part and the system control part. The main components of the power electronic part are the rectifier and the inverter. The rectifier generates a DC voltage supply for feeding the inverter. In the thesis, it is realized both as a controlled and an uncontrolled rectifier. The inverter generates a three-phase AC voltage in order to drive the AC induction motor. As one main contribution of the thesis, the power electronic part is realized such that all relevant signals can be measured and analyzed for educational purposes. The system control part uses the F28035 DSP for computations. V/F control is implemented with the space vector modulation method for open-loop speed control. In addition, closed-loop speed control is performed based on a system plant model, that is found with the help of MATLAB. Finally, a new method for controlling the maximum available torque is realized. The study is supported by various measurement experiments that are obtained using the data acquisition system D-LAB
Radius of curvature of bessel-gaussian beam
(2012) Akkoyun, Sıdıka Türkan
In this thesis, for a turbulent atmosphere, the radius of curvature of Bessel- Gaussian beam is formulated. For various order of Bessel-Gaussian beam of the first kind, the source size, propagation distance, wavelength, this formula is analyzed numerically in moderate turbulence, high turbulence and under free space condition. Results have shown that Bessel-Gaussian beam behaves as Gaussian beam and radius of curvature of Bessel-Gaussian beam decreases with growing turbulence levels. Results have also shown that the radius of curvature increases with the increasing source size and changes slowly with the wavelength.
Public safety communication services requirement and their applications in the Turkish geographic structure
(2008) Usta, Mesut Bedri
Türkiye is a country which may face with many natural and social disasters due to the geography surrounding it. In this context, fast and effectual intrusion to the disasters is vital for the comfort and prosperity of the country. Certainly, such an intrusion demands a widespread, advanced, and cost-effective communications system which provides interoperability among different foundations like army, police, civil defence, fire brigade, ambulance, and some civil society organizations. At this point two systems come to front as the solution: APCO25 widely used in North America and TETRA widely used in Europe. In the studies of this thesis the aforermentioned two systems are compared with respect to their technical specifications especially the coverage areas; the results obtained are investigated in detail; and finally, by depending on a technology selection model in the literature, one of the systems is proposed as a solution. The coverage analyses are implemented using the computer program National Spectrum Management System (NSMS) developed by TÜBİTAK for the base stations of two alternative systems placed on several certain locations in Ankara province and the results are submitted in a visual manner. Analyses are divided into two classes to compare the coverage area of a single base station and the total coverage area of a group of base stations.
VHF data technology in air traffic services
(2006) Bellikli, Ahmet Eren
In this thesis, we have introduced communication, navigation and surveillance system?s working principles and bottlenecks based on working performances used in Air Traffic Control systems. We investigated the possible system solution to meet current bottlenecks, safety issues and Air Traffic growth handling for today and in the future. Air Traffic Navigation continuation and communication quality regarding safety issues are directly dependent on the chosen technology base. Choosing the right technology to meet requirements is what we have searched in this thesis. The baseline technology of Broadband -VHF project is as in the fourth generation (4G) mobile communications. iv B-VHF approaches communication in two different ways The one from ground stations to aircrafts (forward link) is based on combination of OFDM-CDMA called MC-CDMA and the second link from aircraft to ground (reverse link) is based on combination of OFDM-FDMA. We added Code Division Multiplexing into the reverse link to enhance the link robustness against interferences and data security. In order to realize this, we developed a Matlab code to simulate its performances to observe whether this combination provides robustness against interferences. Using the simulation we have obtained BER results at different cyclic prefix, number of OFDM symbols, code length size and channel settings. BER obtained with the inclusion of CDM are slightly favorable as compared to OFDM only reverse link, however, the performance improvement is not within the expected scale
Study and design of mimo - ofdm system operating over wireless channel
(2013) Hammadi, Yousif Ibrahim
In this thesis, a new MIMO-OFDM model has been introduced. The performance of this model using PSK and QAM digital modulations for different antenna configura-tions is analyzed and evaluated under Rayleigh fading channel. MIMO-OFDM sys-tem can be implemented using higher order modulations to achieve large data capaci-ty and remarkable enhanced performance. The main purpose of using high order antenna configuration is to increase the space diversity, which will further decrease the BER at given Eb/N0 as compared to lower order Antenna configurations. It is found effectively the diversity order increases as number of receiving antenna increases regardless the number of transmitting antennas , also the lowest BER can be obtained at highest number of transmitting and receiving antenna configurations. MIMO-OFDM using PSK modulation provide remarkable lower bit error rate as compared to QAM technique at same constellation and antenna configuration.Finally , the effect of channel order on MIMO-OFDM system has been reported.
Performance analysis of detection techniques in direct sequence code division multiple access (DS-CDMA) communication systems
(2004) Alat, Gökçen
DS-CDMA is a new technology providing significant progress in wireless communication. Each user is distinguished by a unique spreading code to modulate an information signal. There are two factors limiting a DS- CDMA system: channel noise which is modelled as an additive white Gaussian noise and multiple access interference due to the many user simultaneously sharing the system. Generally, there are three different methods in the detection of user's signal: the single user detection technique, the optimal detection technique and the suboptimal detection technique. m In previous works, conventional detector and the suboptimal multiuser detection techniques are compared using Gold sequence which is produced from the pseudo noise code sequence. In this thesis, conventional single user detection technique is compared against the decorrelating detector and minimum mean square error (MMSE) based on suboptimal linear multiuser detection techniques. Three types of signature waveforms are used in this study: Gold sequence, small set of Kasami and the large set of Kasami. The small set of Kasami displays best BER performance. The large set of Kasami provides better performance than Gold sequence. Finally, in addition to frequently used Gold code sequence on the detection techniques, this thesis makes a comparison according to the usage of different code sequences that have lower cross correlation property

Browse

Browsing Elektronik ve Haberleşme Mühendisliği Bölümü Tezleri by Language "en"