Elektrik Elektronik Mühendisliği Bölümü Yayın Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/411
Browse
6 results
Search Results
Article Yeni Rotalama Algoritmalarının 802.16j AĞI Etkin Çıktı Oranı Artırımına Düşük Araç Hızları Altındaki Etkileri(2011) Preveze, BarbarosÇoklu ortam verileri içeren kablosuz gezgin ağlarda, yeni bilişsel yöntemler ve rotalama algoritmaları kullanılarak, sistemdeki rota ömrü, bağlantı kesinti miktarı, ortalama sekme sayısı ve paket kaybı gibi performans parametrelerinin iyileştirilmesiyle, IEEE 802.16j ağ yapısının etkin çıktı oranının arttırılması sağlanmıştır. Bu amaçla, mevcut IEEE 802.16j ağında kullanılmakta olan OFDMA ve TDMA erişim tekniklerine ek olarak kullanılmak üzere önerilen, En Çok Sıkışan İlk Erişir (MCAF), Spektrumsal Yardımlaşma (SA) ve Arabellek Yönetimi (BM) metotları ile 802.16j ağında etkin çıktı oranı artırımı için elde edilen simülasyon sonuçları, elde edilen teorik sonuçlarla ve literatürde elde edilmiş olan diğer çalışmaların sonuçlarıyla kıyaslanarak doğrulanmıştır. Bu çalışmada ayrıca, önerilen AEABR (Erişebilirlik Tabanlı Rotalama Alternatif Geliştirimi) ve ATAABR (Erişilebilirlik Tik Ortalamalı Erişebilirlik Tabanlı Rotalama) isimli yeni uzun ömürlü rotalama algoritmalarının, geliştirilen etkin çıktı oranı yükseltimi metotlarıyla birlikte uygulanmasıyla, diğer rotalama algoritmalarına göre, daha da yüksek etkin çıktı oranları elde ettikleri gösterilmiştir. Önerilen yeni metotlar, dağınık ağ yapılarının, gezgin düğümler tarafından, anlık plansız sinyalleşme ile yönetimine dayanarak çalışmaktadır.Article Navigation Under GNSS Denied Environments: Zero Velocity and Zero Turning Update(2022) Çifdalöz, OğuzhanThe objective of this paper is to present a method which bounds the error of an inertial navigation system (INS) when Global Navigation Satellite System (GNSS) is not available. Inertial navigation systems utilize gyroscopes and accelerometers, and calculate velocity, position and attitude, essentially by integrating the measurements obtained from these sensors. Due to the nature of integration, INS are notoriously prone to sensor biases and drifts. Typically, GNSS is used to correct the navigation system errors caused by the inertial sensor measurements. However, in GNSS degraded or denied environments, alternative solutions are required. If the platform on which an INS is mounted is known or estimated to be stationary, zero-velocity update (ZUPT) and/or zero turning update (ZTUPT) algorithms can be applied in order to bound the navigation system errors. Under certain assumptions, ZUPT based algorithms can be applied when the platform is not stationary. If a vehicle’s motion is constrained by the design of its kinematics, i.e. if it can be assumed that the vehicle cannot move or rotate along one or more of its body axes, ZUPT assisted Kalman estimators can be used to correct the errors along those axes. Potentially, ZUPT based estimation algorithms can also be utilized when a sufficiently high fidelity vehicle model is available. In this paper, the implementation of zero-velocity update (ZUPT) and zero turning update (ZTUPT) algorithms are analyzed for the purpose of estimating and bounding inertial navigation errors. The basic principle in navigation is based on combining the data obtained from the sensors onboard and the inertial navigation system through an Extended Kalman filter. Although this process requires additional software components, it potentially offers increased system accuracy and reliability. Incorporating the kinematics of the vehicle, along with a ZUPT and/or ZTUPT algorithm, provides additional data to feed into the Kalman filter and increases the efficiency of error estimation. Estimated error is then fed back into the INS algorithm in order to counteract the sources of error.Article Conducted Emi Performance Comparison Of Si And Sic Mosfets In A Ccm Boost Pfc Converter For MIL-STD-461F CE102(2018) Kavak, Halil; İskender, İres; Jahi, AmirThis paper presents a comparison of conducted EMI performance of Si and SiC MOSFETs in a CCM PFC boost converter that is designed to meet CE102 of MIL-STD461F. EMI performance comparison is based on MOSFET of the PFC converter. That is, the power switch of the converter is the only parameter that is changed during tests. The boost diode is kept the same during the tests and the type of the boost diode is SiC. The paper shows the CE102 test results of Si and SiC MOSFETs without an EMI filter at the input side of CCM PFC boost converter.Article A Novel Analytical Method for Throughput Calculation of Wireless Ad-Hoc Networks Running Different Routing Algorithms(2018) Preveze, BarbarosBecause of the increasing number of internet related applications, the role of total router transmission delay became much more important for the service quality. For this purpose, the tunneling techniques have been widely used especially for real time multimedia transmission to have less number of route constructions and to be able to forward each packet at each router without the need of reaching the upper OSI (Open Systems Interconnection) layers. But, in mobile networks, since the network experience with more changes in traffic conditions and node locations, tunnels will be reconstructed for many times and some extra delay will occur to reconstruct these tunnels. In this work, the place of the tunneling algorithm is taken by the well-known MPLS (Multi-Protocol Label Switching) protocol and for confirmation the throughput calculations are made by considering two different routing algorithms, one of which is AEABR algorithm proposed in [1] (shown in [2] that it improves the system throughput w.r.t Fastest path Routing algorithm [3] for various vehicular velocities), and the other one is Fastest Path routing algorithm [3]. In this work a novel analytical method for throughput calculation of wireless ad-hoc networks running aforementioned routing algorıthms is proposed including the effects of extra delay caused by extra Route Reconstructions (RRC).Article Dynamic Optımızatıon of Image Brıgthness Level With Optimal Gamma Value Assessment (OGVA) Method(2020) Preveze, BarbarosIn this study, the proposed Optimum Gamma Value Assignment (OGVA) method is intended to dynamically optimize the image intensity level in non-desired images due to undesired light levels. For this purpose, it is aimed to make the dark images which cannot be seen due to lack of light, while bright images are dynamically dimmed by using the optimum gamma correction value applied on the image momentarily. It has been shown that this novel method, which will only be implemented as software, without requiring any additional hardware, yields satisfying results even at different light levels.Article Defining Image Memorability Using the Visual Memory Schema(2020) Akagündüz, Erdem; Bors, Adrian G.; Evans, Karla K.Memorability of an image is a characteristic determined by the human observers' ability to remember images they have seen. Yet recent work on image memorability defines it as an intrinsic property that can be obtained independent of the observer. The current study aims to enhance our understanding and prediction of image memorability, improving upon existing approaches by incorporating the properties of cumulative human annotations. We propose a new concept called the Visual Memory Schema (VMS) referring to an organization of image components human observers share when encoding and recognizing images. The concept of VMS is operationalised by asking human observers to define memorable regions of images they were asked to remember during an episodic memory test. We then statistically assess the consistency of VMSs across observers for either correctly or incorrectly recognised images. The associations of the VMSs with eye fixations and saliency are analysed separately as well. Lastly, we adapt various deep learning architectures for the reconstruction and prediction of memorable regions in images and analyse the results when using transfer learning at the outputs of different convolutional network layers.