Theses supervised by Prof. Dr. Hüseyin Canbolat
16 theses · Ankara Yıldırım Beyazıt University
Environmental sound database design and implementation
This thesis proposes "An Environmental Sound Database Design and Implementation" for environmental sound recognition. Environmental sound recognition has become a hot topic in recent years. Environmental sound recognition can be used in security systems, in robot navigation, in internet search engines and more. It is necessary to establish a database which comprises the sounds to be recognized in advance to be able to recognize the environmental sounds. Sound database development is the first stage of all environmental sound recognition tasks. In this study, a database based relational database system is designed.. The sound databases designed and used so far are all constructed on the directory and file system of the operating system. For the first time, a sound database system was constructed using relational database and SQL (Structured Query Language) was used. After database design, the example sounds was loaded, a GUI (Graphical User Interface) as a desktop application and a web interface for accessing over internet was designed.
Longitudinal and lateral-directional static and dynamic stability characteristics evaluation on B60 Duke
The test crew has been assigned to assess the B60 Duke suitability for basic transport training role in terms of Longitudinal and Lateral-Directional Static and Dynamic Stability. The evaluation was performed as test techniques of stabilized method, stabilized turn, pull up, pushover, short period, phugoid, WLSS, SHSS, Dutch Roll,spiral mode and bank-to-bank in B60 with tail number C-FYWQ on 22nd July and 12th, 16th, 19th August 2022. It was assessed as SATISFACTORY that longitudinal static stability, longitudinal maneuver stability, longitudinal dynamic stability (except phugoid mode in climb configuration), lateral-directional static stability and lateral directional dynamic stability of the aircraft were positive. It was evaluated that the B60 was suitable for basic transport training flights without any deficiency. All test points were executed without onboard flight test instrumentation. All recorded data was collected by hand notes. Further flight tests should be executed with onboard flight test instrumentation to increase the accuracy of the test results (Highly Desirable-Mission Effectiveness). Further flight tests in different weights, C.G, airspeeds, altitudes and configurations should be performed to completely evaluate the longitudinal and lateral-directional static and dynamic stability of the B60 Duke (Highly Desirable-Mission Effectiveness). Aircraft met with the CS23.157, CS23.173, CS23.175, CS23.177, CS23.181, MIL-F 8785C 3.2.2.2.1, MIL-F-8785C 3.3.1.1, MIL-F-8785C 3.3.1.3, MIL-F-8785C 3.3.6.1, MIL-F-8785C 3.3.6.2 and MIL-F-8785C 3.3.6.3 requirements. Key words: Stability, longitudinal, lateral, directional, pull up, pushover, short period, Phugoid, WLSS, SHSS, Dutch Roll.
Defensive systems against unmanned vehicles
Unmanned Aerial Vehicles are aerial vehicles that can be remotely controlled or act autonomously and that can be fitted with payloads. Today, UAVs are used in many different areas such as reconnaissance, surveillance, scientific research, firefighting, agriculture and forestry, communication, trade, bomb or missile air attacks and border protection-control. The fact that UAVs are preferred over airplanes are cost effective, have a low level of human factor in flight and can be used in areas with chemical, biological, radiological and nuclear contamination. Due to these advantages, it has been observed that the use of UAVs has increased. However, this increase has begun to be seen as a threat to countries; threfore, many countries have sought a defense system against UAVs. For this reason, in my thesis, general information about the UAVs used for military purposes, which is the subject of the study, was given, the methods of neutralizing the UAVs used for offensive purposes and the defense systems used against them were examined. In the conclusion of all the analyses, in addition to Defense Mechanisms against existing UAVs, the use of Kamikaze Drones has been recommended. The reasons for proposing the use of Kamikaze Drones have been elucidated. Their advantages, such as enabling focused and proportional use of force, providing a more precise and customizable approach, and the ability to be deployed more swiftly, contribute to the recommendation for the use of Kamikaze Drones. It is emphasized that both Air Defense Systems and the use of Kamikaze Drones should not be considered in isolation, but rather approached as a conceptual whole. The operational principle of the system has been explained through the formulation of an algorithm. In light of all this information, the utilization of Drones against threatening UAVs is considered to be a significant force multiplier. Looking at the conclusion, it is assessed that the threat posed by evolving UAV systems being used against nations will increase, underscoring the necessity to enhance defense systems against UAVs in this context. Keywords: UAV, unmanned aerial vehicle, drone, classification, uav attacks, detecting, neutralizing, anti-drone system, anti-uav system, counter uav system
Design and implementation of an image based steganography
Along with tremendous increase of software based applications, information privacy has become one of the overborne issues today for efficiency of electronic data processing. The main purpose on privacy of communication is to provide secure connection with target while avoiding to be captured by the third person in a way that flowing data should not be construed. In well accepted sense, steganography is the study and practice of concealing information to provide security of communication. Various LSB methods derived from classic LSB (least significant bit) are used as steganography techniques on spatial domains. In case of direct use of the classic LSB technique, security concern becomes an issue because of its simplicity and predictability. Also, traditional applications of classic LSB inevitably causes distortion of image. In this thesis, carried study proposes an improved LSB Steganography technique to enhance security while decreasing the rate of distortion in subject image. Proposed method in this study is based on hiding two bits of secret data in one color (RGB) pixel with only one least significant bit change in one of the layers. LSB value of Red layer is utilized in XOR operation with both Green and Red LSB values respectively. By applying this technique, only with one LSB bit change, two bits of secret data can be hidden as a consequence of XOR bitwise operation nature. XOR based operation provides more secure and unpredictable communication when compared to classic LSB. Aforementioned methods applied on different images, and it was found that improved method studied here revealed favorable results for the concerns of Peak Signal to Noise Ratio (PSNR) and Mean Square Error (MSE) when compared to the classic LSB method.
Design and analysis of multiband compact antennas for future 5g applications
Haberleşme sistemlerinde ileride çıkabilecek 5G uygulamalarında kullanılabilecek çok bantlı kompakt antenlerin yeni modellerinin tasarım ve analizi bu çalışmada sunulmaktadır. Bu çalışmadaki ana yaklaşım ölçek küçültme ve yarık (slot) azaltma işlemlerine altyapı sağladığından üç bantlı mikro-şerit yama anten yapısına dayalıdır. Önerilen antenin özelliklerini ve eklenmiş mikro şerit iletim hattı besleyicisi gibi doğrudan besleme yöntemini iyileştirmeyi hedefleyen çeşitli biçimlerde yarıkları olan yamalardan oluşmuş anten yapıları iyi bir bant genişliği empedans uyumlaması elde etmek için kullanılmıştır. Bu bağlamda, anten dizaynında kullanılan çift bant yaklaşımı yama içindeki elektrik akım davranışı yardımıyla çeşitli anten yapılarında değişiklik yaparak elde edilebilir. Bu hedefe erişmek konusunda yamanın ve toprak plakasının boyutlarını değiştirme işleminin çok etkili olduğu bulunmuştur. Çift bantlı antenler Kabul edilebilir bir yayınım ve kazanç özelliği yanında amaca uygun bir çift bant çıkışı almak için çift bantlı antenler imal edildi. Bu tezde, gerekli çift bant rezonans frekanslarını üretmek ve istenmeyen frekansları söndürmek için farklı şekillerde oyulmuş yeni yarıklar eklenmiş üç tane çok bantlı mikro şerit kompakt anten önerilmektedir. Önerilen çok bantlı antenlerin hepsi 50 Ohmluk içeriye beslemeye bağlanmış 0,508 mm kalınlıkta 2,2 bağıl geçirgenlikli Roger RT5880 Duroid altlığı üzerine yerleştirilmiştir. Bu antenlerin performansları incelenmiş, analiz edilmiş ve çeşitli parametreler açısından değerlendirilmiştir. Antenlerin modelleme ve performans değerlendirme işleri en yaygın EM simülatörlerinden biri olan Computer Simulation Technology Microwave Studio (CST MWS) program paketi ile yapılmıştır.viii Bu antenler daha önceki araştırmalarda tasarlanan antenlere ve bu çalışmada ele alınan referans antene (8,98 mm2) göre daha küçük ölçeklerdedir. Önerilen bütün tasarımlar 10 GHz X-band ve 30 GHz Ka-bantta çalışabilir.
Zirai ilaçların tarım çalışanları üzerine olumsuz etkileri ve bu etkileri azaltma tedbirleri
Değişik iklim ve toprak çeşitliliğine sahip olan ülkemizde buna bağlı olarak zengin bir tarımsal ürün çeşitliliği görülmektedir. Topraklarımızda yetiştirilen ve ekonomik önemi olan birçok bitkinin elde edilmesinde çok çeşitli zararlılarla mücadele edilmesi gerekliliği bilinmektedir. Ürün kayıplarına neden olan ve verimliliği düşüren zararlılar, hastalıklar ve yabancı otlar ile yapılan mücadelede en yaygın ve en pratik yöntem bitki koruma ürünlerinin kullanımıdır. Tarımda yapılan çalışmalarda ortaya çıkan birçok sağlık sorunu, yaygın olarak kullanılan bitki koruma ürünlerine maruziyet ile ilişkilidir. Bu sağlık sorunları arasında akut veya kronik olarak ortaya çıkabilen; kanserojen, immünolojik, nörotoksik ve üreme etkileri bulunmaktadır. Bu etkileri düşünüldüğünde, zirai ilaçların (bitki koruma ürünlerinin) kullanımının çalışanlar üzerine etkilerinin ve bu etkilerin önlenmesi için alınması gerekli tedbirlerin kapsamlı bir şekilde değerlendirilmesi gerektiği açıktır. Ülkemizde tarım çalışanlarının büyük çoğunluğu 6331 sayılı İş Sağlığı ve Güvenliği Kanunu ile 2012 yılında iş sağlığı ve güvenliği mevzuatı kapsamına alınmıştır, ancak halen tarım sektöründe zirai ilaç kullanımına yönelik olarak detaylı bir alt mevzuat bulunmamaktadır. Bu tez ile tarım sektöründe zirai ilaç kullanımında alınacak tüm önlemlere yönelik bir kaynak oluşturulması amaçlanmıştır. Bu tez çalışmasında ile öncelikle genel olarak zirai ilaçların etkileri ve zirai ilaçlarla çalışmalarda alınacak önlemlere yönelik olarak kapsamlı bir literatür ve mevzuat araştırması yapılmıştır. Yapılan araştırma sonucunda elde edilen bilgiler sonucunda zirai ilaçlarla çalışmaların tüm aşamalarında alınacak tedbirler ile iyi uygulamalar detaylı olarak verilmiştir. Tez çalışmasının sonucunda ise tarım sektöründeki çalışmalarda kullanılmak üzere zirai ilaçlara maruziyetin önlenmesine yönelik olan ve çalışmalar sırasında kullanıldığında maruziyeti önemli ölçüde azaltacak bir kontrol listesi oluşturularak sunulmuştur. Anahtar Kelimeler: İş Sağlığı ve Güvenliği, Pestisitler, Zirai ilaçlar.
Farklı çalışma ortamlarında kullanılan yapay aydınlatma kaynaklarının spektral analizi ve çalışanlar üzerindeki etkileri
Çalışma hayatı şartlarını düşündüğümüzde, yapay ışık kaynaklarına olan maruziyetimizin anlık ve kümülatif etkilerinin çalışanlar üzerinde bırakacağı etkinin kaçınılmazlığı gerçeğini düşünmekte, teknolojik gelişmeler ile birlikte bu ürünlerin sadece, çalışma ortamlarını aydınlatmakla kalmayıp, farklı formatlarda hayatımıza girmiş olmaları, aslında sadece çalışan olarak değil toplum içerisinde yer alan her birey için farklı çıktıları olduğu değerlendirilmektedir. Enerji verimliliği konusunun üst düzeyde tartışıldığı, yeni aydınlatma teknolojilerinin denendiği ve piyasaya sürüldüğü bu ortamda, bu teknolojiden kaynaklanan foto biyolojik, fotokimyasal ve foto termal reaksiyonların hem göz sağlığı hem de cilt sağlığı açısından, akut ve uzun süreli sonuçlar doğuracağı, yapılan bilimsel çalışmalar ve standartlarda yer almaktadır. Bu anlamda, yapacağımız çalışma ile genel bir çerçeve içerisinde çalışma hayatında kullanılan yapay ışık kaynaklarının bir kısmının spektral analizlerinin yapılması ve piyasada bulunan ürünlerin gerçekten de istenen değerlere ulaşıp ulaşmadığının ve bunun çalışanların sağlığına olan etkilerini incelemek olacaktır. Yapay ışık kaynaklarının sınıflandırılması, sınıflandırılan bu kaynakların her birinin beklenen spektral dağılım grafiklerinin paylaşılması ve numune olarak seçilen kaynakların spektral analiz ölçümlerinin yapılarak karşılaştırılması ve değerlendirme sonuçlarının hangi sonuçlar doğuracağının paylaşılması olacaktır. Maruziyet sonucu oluşan zararların ve spektral band genişliği değerlerinin, risk grupları ölçeğinde değerlendirilmesi ile bu durumları önlemeye yönelik çıktılara ulaşmak ana hedefimiz olacaktır.
TDM data link design and optimization between unmanned aerial vehicle and tactical guided munition
Unmanned aerial vehicles (UAVs), a creation of modern technology, are becoming armed and seeing an increase in use in military missions. And the nations that possess these UAV systems benefit strategically. Undoubtedly, tactical guided ammunition used with UAVs is becoming more significant in the defense industry in addition to UAVs. The study's scope includes an explanation of the munitions' guidance systems as well as the subsystem components that both UAVs and tactical guided munitions are made up of. The primary goal of the study is to design and optimize a data link—the wireless, digital communication between tactical guided munitions and unmanned aerial vehicles—by taking into account its properties and characteristics. The "Man in the Loop" concept incorporates this data link, which is future technology. In the study's application section, designs are produced at various data rates by setting up design input parameters and selecting various digital modulation techniques. MATLAB codes are embedded into the Modefrontier optimization program during the optimization phase, and the MOGA (Multi-Objective Genetic Algorithm) genetic optimization technique is used to generate a set of designs consisting of 400 different optimal designs. The designs created in the application part are evaluated to determine if they belong in the collection of Modefrontier's top designs. Also, the impacts of particular communication modulation techniques on designs are looked at. Keywords: UAV, Tactical Guided Munition, Data Link, Optimization, Digital Communications
Performance analysis of optical wireless communications system in Ankara
From the depth of ancient history, humans were searching for the science of telecommunication. The Chinese used colorful flags and drums, and the Arabs used pigeons. There was an urgent need for science to find quick and effective means of communication. Modern communications were a technical revolution in the eighteenth century and the use of light was one of the pillars of this technical revolution. But in fact, the use of light for communication was much older than we know in books. In modern communications, the light was used as a means of transmitting data in World War II aboard the British Navy. And its military uses in communications are many. Today, light is used in fiber optics as a signal carrier, and it is considered one of the fastest means of communication, but it is flawed by its monopoly by the state and the cost of constructing its expensive channels, drilling, and costly maintenance. On the other hand, there is a system called FSO, which works similarly to the idea of sending data through light in a fiber optic. The main goal of this dissertation is to analyze the Free space optical system with the weather condition of Ankara, the capital city of turkey. Governmental meteorological data support the analysis of this study of wireless communication systems; where it provides real measurement values. This system is used in short-range applications, which are between buildings in cities or between skyscrapers, and sometimes between ships, as well as very complex military situations. This study will investigate the FSO using a semiconductor laser according to the weather conditions of Ankara and the distance traveled taking into account a good high-quality data transmission.
Environmental sound recognition with various feature extraction and classification techniques
This thesis proposes "An Environmental Sound Recognition with various feature extraction and classification techniques" for environmental sound recognition. Study in Environmental Sound Recognition (ESR) has taken attention in recent years. In the past decade, research on the Environmental Sound Recognition (ESR) area has accelerated. ESR has important role on intelligent computer systems and robots for the purpose of identification, recognition and discrimination. In this survey, I will put forward a survey on which various feature extraction and classification techniques is better to recognize environmental sounds. Survey includes these parts: environmental sound recognition system processing, feature extraction techniques, classification techniques, and performance comparison of selected techniques. At long last, finishing up comments and future innovative work slants in the ESR field will be given.
Missile telemetry system design and performance analysis
Telemetry is either wired or wireless remote monitoring or control of a system. It is also used in the sense of 'receiving remote measurement' as a word meaning. Thanks to the telemetry system, various data belonging to a system are collected from a certain distance and this data is transmitted to a point to be displayed. Telemetry system is used in many application areas. Within the scope of this study, the focus is on the missile telemetry system. During the development phase of the missiles, various External Store Certification tests are carried out. These tests are very important for collecting data and determining the working limits for the design of the weapon system and its subcomponents under development. The more ground and flight tests performed, the more accurately the design works. A telemetry system is needed at the stage of collecting this data. Thanks to the test head part placed instead of the warhead, data is collected from the electronic card and various sensors in the missile. In this study, the missile telemetry system design and the Telemetry System Performance Calculator Interface program developed in line with this design are emphasized. Thanks to this interface, Link Margin of the Signal to Noise Ratio value is calculated by taking into account the Radio Frequency communication values used in the telemetry system on both the transmitting and receiving sides and the modulation types. According to these modulation types, telemetry communication performances are compared. As a result of the study, Link Margin of the Signal to Noise Ratio in Binary Phase Shift Keying and Quadrature Phase Shift Keying modulations are better than other modulations. As a result of these studies, we provide a more reliable environment for pre-flight by pre-determining the communication parameters of the telemetry equipment required for missile telemetry system design. Keywords: Telemetry, missile telemetry system, modulation, Link Margin, Telemetry System Performance Calculator Interface
Infrastructure-less illumination and communication networks
Visible light communication VLC is a wireless technique that works in combination with well-known radio frequency (RF) communication and has the potential to deal with the spectrum crunch issue in increasingly crowded RF spectrum regions. The knowledge of physical layer approaches for visible light point-to-point links has advanced significantly in recent years, but the fundamental issue of creating effective networking technology tailored for visible-light networks remains largely unsolved. The methodologies now in use as well as the major difficulties in designing visible-light ad hoc networks, or LANETs, are reviewed in this study. VLC offers a number of difficulties, including interference and obstruction-caused blockages. Various settings also provide a wealth of light sources that may be employed to support end-to-end communication.
Control and comparison of an electromechanically actuated control surface according to different sensor locations
In missile systems with a control actuation system, in terms of control methods, sensors and their locations have an important effect on control methods. Transmission mechanisms between the sensor measuring locations and control surfaces make control of system difficult due to their non-linear behavior. This nonlinear situation is caused by backlash and deformations between control surfaces and sensor position. In this study, differences between direct feedback and indirect feedback and differences caused by these differences in terms of controller will be examined. In order to study this situation, position sensors are placed on two different mechanism parts for an electromechanical fin actuation system, which is responsible for directing a missile. After this placement, system identification of the fin actuation system has been made with measurement from fin angle. After that, controllers have been designed. The first controller was designed by method of state feedback. In first simulation studies, it is observed that the controller made steady state errors under the disturbing effect. In order to eliminate the steady state error, an integrally effective outer loop has been added to the controller that the error resets. As an alternative to the controller designed with state feedback method, a PID controller with same roots has been designed. Then the controllers are in real time tested. In order to minimize the effects of measurement errors and nonlinearities, backlash and material deformations caused by the nature of the mechanism by using the feedback from two different sensors, it has been observed that the system response is cleared from the error values caused by the backlash and the system can be controlled more accurately. Keywords: Electromechanical, Sensor, Control
Detection and classification of forest fire smoke by deep learning methods using Unmanned Aerial Vehicle in Türkiye
It has been observed that forest fires have increased over the years in the world and in Türkiye. This disaster, which emerged as a result of various events such as climate change, human factors or lightning strikes, caused serious damage to the ecosystem and started to pose great dangers. Direct or indirect destruction not only affects animal and plant ecosystems, but also affects many areas such as climate change, tourism, industry and health in the long run. There are many ways to respond to forest fires. Thanks to today's technologies, early intervention in fires has become easier and more systematic than before. Thanks to deep learning algorithms, the chance of responding to fires increases considerably. In this study, we made classification by estimating whether the fire was an old or new fire, thanks to the images taken from the videos taken on the Unmanned Aerial Vehicle, and an accuracy rate output was obtained with deep learning algorithms. Since the noise values of these pictures taken from motion cameras are high, their loss values are higher than other still cameras. For this reason, by creating a model with deep learning algorithms and increasing the data, 96.97% accuracy value was obtained in our test data. As we can see from the result here, it has been observed that the models created from deep learning algorithms are successful by reaching high accuracy rates. Since the noise values of these pictures taken from motion cameras are high, their loss values are higher than other still cameras. For this reason, by creating a model with deep learning algorithms and increasing the data, 96.97% accuracy value was obtained in our test data. As we can see from the result here, it has been observed that the models created from deep learning algorithms are successful by reaching high accuracy rates. Since the noise values of these pictures taken from motion cameras are high, their loss values are higher than other still cameras. For this reason, by creating a model with deep learning algorithms and increasing the data, 96.97% accuracy value was obtained in our test data. As we can see from the result here, it has been observed that the models created from deep learning algorithms are successful by reaching high accuracy rates.
Design of helicopter DC power distribution unit
In our modern time, all helicopters have the electrical power to supply the equipment for several systems both flight critical and operational. Helicopter may be used for SAR (Search and Rescue), combat, military tactical, cargo transportation, health (ambulance), farming and commercial, military or VIP transportation. For all usages of modern helicopters have to use electrical power for systems that are used for several missions. The Electrical Power Generation and Distribution System (EPGDS) generate and distribute electrical power to not only the aircraft electrical power utilization equipment but also enable the engine starting. There are two types for electrical power as AC and DC. AC system is mainly used for systems, which has high power consumption (generally supplied from 115VAC, 400 Hz power source) such as ice protection, window-cabin heating, main-tail rotor heating or weapon systems. DC power is used especially on the avionics equipment that are generally supplied by 28VDC and 28VDC power source. This thesis is about the design principles of a helicopter's DC power distribution unit by considering not only the electrical standards but also all the subjects that are necessary to design the power distribution unit (PDU) beginning with requirements till the end of design by using systems engineering principles. There are some assumptions PDU shall be capable of deliver DC electrical power from generators to secondary distribution units (circuit breaker panel, relay panel etc.). Keywords: Helicopter, DC Power, Electrical, Distribution, EPD, PDU
Development of an end-to-end sign language hand gesture recognition system using novel deep learning methods
Sign languages serve as the primary communication tools for individuals with hearing impairments. Developing systems to support these languages is essential due to the challenges such individuals face when interacting with people who are not proficient in sign language (SL). To address this, the present study leverages advancements in Artificial Intelligence (AI). The research involves four integrated phases to develop a Human–Computer Interaction-Based Sign Language Recognition System (HCISLRS). These phases include designing a novel graphical user interface for Sign Language System (G-SLS) with a proposed segmentation method, creating a new dataset for Turkish Sign Language Alphabet (TSLA), developing innovative learning methods, and testing the system in real time. G-SLS uses combined segmentation techniques to detect objects, specifically hands, in images. This method, employed in G-SLS, supports both dataset creation and hand gesture recognition. Following this, a dataset comprising visual samples of TSLA is compiled. The dataset is imbalanced, introducing variations in class distribution, gesture types, backgrounds, user-to-camera distances, and lighting conditions. These variations contribute to the dataset's complexity and enhance its relevance for real-world applications. Subsequently, this research also develops novel deep learning methods: TSLA networks (TSLAnets) and Vision Transformer of Turkish Sign Language Alphabet (ViTSLA). These methods demonstrate high performance across multiple evaluation metrics on various dataset classes. Moreover, a comprehensive comparison is conducted between the different developed methods, established deep learning architectures, and models from previous studies. To assess the generalization ability of the proposed methods, they are also evaluated on external datasets. This ensures that the performance is not dataset-specific but consistent across different data distributions. These comparisons affirm the effectiveness of the innovative methods, demonstrating superior performance measures, faster computation, and fewer parameters compared to other models.