TOBB ETU GCRIS Database
Not a member yet
9778 research outputs found
Sort by
A Novel Approximation Method for Computing the Adjustment Coefficient of a Nonlinear Cramér-Lundberg Risk Model with Gamma Claims
No full textThis study considers a non-linear Cram ; eacute;r-Lundberg risk model and examines the adjustment coefficient (r)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}\varvec{(r)}\end{document} when the claims have gamma distribution. The linear models are not always adequate because an insurance company's premium income does not always increase linearly. Therefore, in this study, a more realistic non-linear Cram ; eacute;r-Lundberg risk model is mathematically constructed. Then, the ruin probability of this non-linear risk model is studied when the premium function is in the form of square root function, i.e., p(t)=ct\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}\varvec{p}\varvec{(t)}\varvec{=}\varvec{c}\varvec{\sqrt{t}}\end{document}. It leads to analyzing the adjustment coefficient (r)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}\varvec{(r)}\end{document}, as examining this coefficient is required for finding an upper bound while investigating the ruin probability. However, in general case, it is a challenging procedure to calculate the exact value of r\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}\varvec{r}\end{document} from an integral equation. Thus, in this study, the adjustment coefficient r\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}\varvec{r}\end{document} is explored by computational methods and a new approximate formula for the practical calculation of the adjustment coefficient is proposed. Moreover, an implementation of the obtained approximate formula, which investigates ruin probability, is included as an example at the end of the paper
Modeling a Humanitarian-Aid Covering Tour Problem with Location Selection and Vehicle Assignment Decisions
No full textIn post-disaster situations, time-critical decision-making is essential. Due to high demand and limited resources, visiting all affected locations is often infeasible. This study presents a novel variant of the covering tour problem that addresses these constraints by integrating location selection and vehicle assignment decisions. In the proposed problem, vehicles depart from selected relief centers, visit a subset of victim locations, and are allowed to complete their tours at any center. The demands of unvisited locations are satisfied through demand transfers from nearby visited nodes, with associated transfer times included in the total operation time. A mixed integer linear programming (MILP) model is formulated to minimize total operation time, incorporating both travel and demand transfer times. Scenario-based analyses are performed to evaluate the model's performance under various operational conditions, including transfer time sensitivity, route flexibility, demand coverage constraints, time-based covering radius, and partial fulfillment policies. To address scalability, a two-stage clustering-based heuristic is developed, offering a practical and computationally efficient solution method. From a humanitarian logistics perspective, the findings emphasize the importance of flexible routing, strategic placement of relief centers, and careful management of coverage thresholds. Additionally, the simplicity and adaptability of the proposed heuristic make it well suited for real-time decision-making in post-disaster response operations
Psychological Impact of Horn Sounds and Headlight Flashing on Novice Drivers During Driving
No full textBu çalışma, acemi sürücülerin işitsel ve görsel uyaranlara, özellikle korna sesi ve selektöre verdikleri psikolojik ve davranışsal tepkileri incelemektedir. Bu çalışmanın temel amacı, acemi sürücülerin korna sesi ve selektör gibi çevresel uyaranlara verdikleri duygusal tepkilerin sürüş davranışlarına olan etkisini incelemektir. 30 acemi sürücü ile yüz yüze veya internet üzerinden görüşmeler yoluyla nitel veri toplanmış ve bu süreçte, acemi sürücülerin trafikte karşılaştıkları uyaranlara verdikleri tepkiler ile karar alma süreçleri ve genel sürüş performansları üzerindeki etkileri incelenmiştir. Katılımcılardan elde edilen veriler tematik analiz yöntemi kullanılarak beş ana kategori altında gruplandırılmıştır: “Çevresel Faktörler,” “Duygusal Tepkiler,” “Sürücü Davranışı,” “İletişim” ve “Tutumlar.” Bulgular, korna sesi ve selektör gibi dış uyaranların acemi sürücülerin dikkatini ve duygusal durumunu önemli ölçüde etkilediğini göstermektedir. Bu durum genellikle kaygı, öfke ve stres seviyelerinin artmasına yol açmaktadır. Bu duygusal tepkiler, hız yapma, agresif sollama ve şerit değiştirme gibi saldırgan sürüş davranışlarıyla yakından ilişkilidir. Ayrıca, acemi sürücülerin bu tür uyaranlara karşı tutumlarının zamanla değiştiği, deneyim kazandıkça duygusal tepkilerinin azaldığı tespit edilmiştir. Ancak, bu uyaranlara sık maruz kalmak bazı durumlarda sürücülerin özgüveninin azalmasına neden olabilmektedir. Çalışmanın sonuçları, sürücü eğitim programlarına duygu düzenleme ve stres yönetimi stratejilerinin dahil edilmesinin önemini vurgulamaktadır. Acemi sürücülerin duygusal tepkilerini kontrol edebilme becerisi kazanmaları, yol güvenliğinin artırılmasına, saldırgan sürüş davranışlarının azaltılmasına ve genel sürüş performansının iyileştirilmesine katkı sağlayabilir
Energy Scale and Resolution for Anti-kt Jets with Radius Parameters R=0.2 and 0.6 Measured in proton-proton Collisions at S=13 TeV with the ATLAS Detector
No full textJets with different radius parameters R are an important tool for probing quantum chromodynamics processes at different angular scales. Jets with small R=0.2 are instrumental in measurements of the substructure of large-R jets resulting from collimated hadronic decays of energetic W, Z, and Higgs bosons, top quarks, and of potential new resonances. This paper presents measurements of the energy scale, resolution, and associated uncertainties of jets with radius parameters R=0.2 and 0.6, obtained using the ATLAS detector. The results are based on 37fb-1 of proton–proton collision data from the Large Hadron Collider at a centre-of-mass energy of s=13 TeV. A new in situ method for measuring jet energy scale differences between data and Monte Carlo simulations is presented. The systematic uncertainties in the jet energy scale for central jets (|η|1.2) typically vary from 1% to about 5% as a function of |η| at very low transverse momentum, pT, of around 20 GeV for both R=0.2 and 0.6 jets. The relative energy resolution ranges from (35±6)% at pT=20 GeV to (6±0.5)% at pT=300 GeV for central R=0.2 jets, and is found to be slightly worse for R=0.6 jets. Finally, the effect of close-by hadronic activity on the jet energy scale is investigated and is found to be well modelled by the ATLAS Monte Carlo simulations. © 2025 Elsevier B.V., All rights reserved
SPECT Sol Ventrikül Bölütlenmesi Kullanılarak Makine Öğrenmesi Yöntemleri ile Kalbin Uzun Ekseni Çıkarım
No full textIsik UniversitySegmentation of myocardial tissue in SPECT (Single Photon Emission Computed Tomography) left ventricle images is a crucial problem for assisting diagnosis. Non-AI-based segmentation models in this field typically segment by first extracting key points such as the apex and base of the heart, as well as lines like the long axis, based on predefined assumptions. However, these models perform poorly in cases where SPECT images are noisy. In contrast, AI-based models, which are more robust to noise, can perform segmentation without requiring any predefined points or axes. In clinical practice, segmented heart images are examined by experts using short-axis, vertical long-axis, and horizontal long-axis views. Therefore, determining the long axis of the left ventricle is of critical importance. As a novel contribution to the literature, this study aims to extract the long axis from binary segmentation images in AI-supported SPECT left ventricle segmentation models - a missing aspect in current approaches. The deep learning model we developed determines the symmetry axis in given 3D binary segmentation images and extracts clinically important cross-sections from long and short axes for diagnostic evaluation. © 2025 Elsevier B.V., All rights reserved
Predictive Beamforming With Distributed MIMO and NLOS Identification
No full textBrowsy; CIS Arge; PTT TeknolojiRadar-assisted beamforming in millimeter-wave (mmWave) massive MIMO (multiple-input multiple-output) significantly reduces beam tracking overhead, making it particularly beneficial for communication between road-side units (RSUs) and vehicles. Most existing studies assume a single RSU and line-of-sight (LoS) conditions. In this work, we consider multiple RSUs performing mmWave distributed MIMO transmission, where temporary non-line-of-sight (NLoS) conditions arise due to obstructions and blockages. Numerical results demonstrate that combining Bayesian NLoS detection with extended Kalman filtering (EKF) in a distributed MIMO framework ensures nearly uniform coverage and high data rate performance. © 2025 Elsevier B.V., All rights reserved
Joint Deep Learning and Atmospheric Light Scattering for Fast Image Dehazing
No full textHaze significantly degrades image and video quality by reducing contrast and visibility. A widely adopted solution is the Atmospheric Light Scattering (ALS) model, which requires estimating two key unknowns: atmospheric light and the transmission map. To estimate the depth map, we employ a lightweight encoder-decoder deep model and, rather than predicting a single air light value for each image channel, we generate an air light map. By this way, for each pixel of the hazy image we can model the real-world hazing effect more properly. This approach leverages the power of both Deep Learning (DL) and traditional methods. Our method decreases the processing load of depth estimation by using a lightweight model. In addition, it expands the air light estimation to all image pixels. This allows our method to differentiate itself from existing works. By keeping the visual quality of the dehazed imagery, our method, on a single GPU, achieves 10 fps and 17 fps for 480p and 360p image resolutions, respectively. Benchmark comparisons demonstrate that this method is competitive with existing state-of-the-art real-time dehazing methods in terms of efficiency and visual quality
What is the Optimal Shape and Size of a Cortical Window That Least Affects Bone Strength? A Biomechanical Study
No full textBackground: Biopsy and curettage are frequently used in orthopaedic oncology surgeries. However, these procedures reduce bone strength and result in pathological fractures. Therefore, meticulous planning of the size and shape of the cortical window is paramount to preserve bone strength. However, few studies have examined the shape and size of the windows created for biopsy and curettage procedures. The purpose of this study was to evaluate the effect of cortical window shape (rectangular versus elliptical) on bone strength and to evaluate the effect of ellipses with different dimensions on bone strength. Hypothesis: The hypothesis of this study was that an elliptical cortical window would have a lesser impact on bone strength than a rectangular window, particularly in cases where the length of the ellipse increases while its width decreases. Materials and methods: Sixty-four synthetic femur models were divided into four groups (n = 8) for compression and torsional tests. Four equal-area cortical windows were created. G1 (rectangular), G2 (elliptical, 1/1.5 ratio), G3 (elliptical, 1/3 ratio), and G4 (elliptical, 1/6 ratio). Compression tests (10 mm/min) were used to assess the maximum load, stiffness, yield load, and fracture energy. Torsion tests (50º/min) were used to measure the maximum torque, stiffness, and work done. Biomechanical performance was compared using load-displacement and torque-angular displacement data. Results: All bone models were fractured along the cortical window. G3 (elliptical, 511.09 ± 55.07 N) had higher maximum load than G1 (rectangular, 389.18 ± 88.46 N, p = 0.003). The elliptical groups (G2:71.21 N/mm, G3:71.04 N/mm, G4:84.10 N/mm) showed greater compression stiffness than G1 (52.60 N/mm, p ≤ 0.05). G3 had higher yield load (458.72 ± 43.42 N) and work done (2.6 J) than G1 (352.43 ± 91.24 N, 1.78 J, p ≤ 0.05). No significant difference in torsion test results was observed between G3 and G1 (p > 0.05). G2 exhibited lower maximum load, yield load, and work than G3 and G4 (p ≤ 0.003). G4 exhibited higher torque (17.08 Nm) than G2 (11.73 Nm, p = 0.02) and G3 (12.62 Nm, p = 0.018). The torsional stiffness was similar across the elliptical groups. Discussion: This biomechanical study demonstrated that elliptical cortical windows, especially those with higher length-to-width ratios, outperform rectangular windows in terms of strength under compression and torsional loads. These findings suggest that elliptical cortical windows may provide biomechanical advantages in terms of preserving bone strength and reducing fracture risk. However, as these results are based on an experimental model, further in vivo studies are needed to confirm their clinical applicability. Level of evidence: V; Comparative laboratory study. © 2025 Elsevier B.V., All rights reserved
Cell-Free Massive MIMO-OFDM With Low-Resolution ADCs
No full textCell-free massive MIMO (multiple-input multiple-output) is a promising infrastructure for 6G and beyond, offering significantly higher spectral efficiency than traditional cellular systems. In cell-free massive MIMO, a large number of low-cost access points (APs) are densely deployed, making hardware impairments inevitable due to cost-effective radio hardware. While the impact of quantization and other impairments has been extensively studied for narrowband channels, their effects in wideband scenarios remain relatively unexamined. This paper presents the first analysis of how low-resolution analog-to-digital converters (ADCs) affect the uplink performance of a cell-free massive MIMO system using an orthogonal frequency division multiplexing (OFDM) waveform. Both quantization-impaired channel estimation and data detection are considered, and the quantization-unaware and quantization-aware linear receivers are developed. To further mitigate the adverse effects of quantization at the bit level, an alternating direction method of multipliers (ADMM)-based receiver is proposed. Simulation results demonstrate that the ADMM-based receiver outperforms conventional linear receivers by orders of magnitude. © 2025 Elsevier B.V., All rights reserved
Yonga Üstü Uygulamalar için Tümleşik Fotonik Izgara Bağdaştırıcıların Tasarımı ve Eniyilenmesi
No full textTümleşik fotonik, optik bileşenlerin tek bir çip üzerinde bütünleştirilmesine olanak tanıyan devrim niteliğinde bir teknoloji olarak öne çıkmaktadır. Bu ilerleme, telekomünikasyon, biyomedikal algılama, kuantum bilişim ve hassas ölçüm sistemleri gibi birçok alanda önemli gelişmelere katkı sağlamıştır. Geleneksel elektronik devrelerden farklı olarak, tümleşik fotonik ışığı bilgi taşıma ve işleme amacıyla kullanarak geniş bant genişliği, ultra yüksek veri aktarım hızı ve düşük güç tüketimi gibi avantajlar sunmaktadır. Ancak, tümleşik fotonik sistemlerde karşılaşılan en büyük zorluklardan biri, optik fiberler ile çip üzerindeki dalga kılavuzları arasında ışığın verimli bir şekilde bağdaştırılmasıdır. Optik iletişim ve algılama uygulamalarında kullanılan standart optik fiberlerin çekirdek çapı yaklaşık 10 μm iken, çip üzerindeki fotonik dalga kılavuzları genellikle 1 μm'den daha küçük boyutlara sahiptir. İki arayüz arasındaki bu önemli boyut farkı, bağdaştırma kayıplarına ve mod uyumsuzluklarına neden olarak tümleşik fotonik cihazların verimliliğini doğrudan etkilemektedir. Bu sorunu çözmek amacıyla çeşitli bağdaştırıcı teknikleri geliştirilmiş olup, en yaygın kullanılan yöntemler ızgara bağdaştırıcılar (grating coupler) ve kenar bağdaştırıcılardır (edge coupler). Izgara bağdaştırıcılar, gelen ışığı kırınım temelli bir yapı aracılığıyla dalga kılavuzuna yönlendirerek dikey bağdaştırma mekanizması sağlar. Bu yöntem, hizalama toleranslarının daha esnek olması, yonga (wafer) seviyesinde test edilebilmesi ve kompakt bir tasarıma sahip olması gibi avantajlar sunarak yüksek yoğunluklu tümleşik fotonik sistemler için cazip bir seçenek haline gelmektedir. Kenar bağdaştırıcıların bağdaştırma verimlilikleri daha yüksek olmasına rağmen, yüksek hassasiyet gerektiren üretim süreçleri ve yonga üstüne ışığı hizalamadaki hassasiyetin önemi gibi zorluklar nedeniyle uygulamada daha karmaşık hale gelmektedir. Tümleşik fotonik platformlarında en yaygın kullanılan malzemeler silisyum (Si) ve silisyum nitrür (SiN)'dür. Yüksek kırılma indisi kontrastına sahip silisyum fotoniği, güçlü optik alan hapsi sağlarken, CMOS (Bütünleyici Metal Oksit Yarı İletken) üretim süreçleri ile tam uyumluluğu sayesinde büyük ölçekli entegrasyon için uygun bir platform sunar. SiN ise düşük yayılım kayıpları ve geniş spektral çalışma aralığı sayesinde özellikle doğrusal olmayan optik uygulamalar ve kuantum bilişim için avantaj sağlamaktadır. Literatürdeki çalışmalar çoğunlukla tek bir malzemeye odaklanırken, bu tezde hem Si hem de SiN kullanılarak 1550 nm dalga boyu için optimize edilmiş ızgara bağdaştırıcı tasarımları önerilmektedir. Bu çalışma, ızgara periyodu, aşındırma derinliği ve bağdaştırma açısı gibi yapısal parametreleri optimize ederek bağdaştırma verimliliğini artırmayı ve kompakt bir tasarım sağlamayı hedeflemektedir. Önerilen tasarımlar, fiberden çip üzerine ışık bağdaştırma sürecinin performansını iyileştirerek tümleşik fotonik teknolojilerinin daha geniş bir uygulama alanına yayılmasına katkıda bulunacaktır.Integrated photonics has emerged as a transformative technology, enabling the monolithic integration and large-scale integration of optical components on a single chip. This advancement has facilitated significant progress in fields such as telecommunications, biomedical sensing, quantum computing, and precision metrology. Unlike traditional electronic circuits, integrated photonics leverages light for information transmission and processing, offering distinct advantages such as high bandwidth, ultra-fast data transfer, and low power consumption. However, one of the major challenges in integrated photonics is the efficient coupling of light between optical fibers and on-chip waveguides. In optical communication and sensing applications, standard optical fibers used for light transmission typically have core diameters of around 10 μm, whereas photonic waveguides on a chip can have dimensions smaller than 1 μm. This significant mode size mismatch results in substantial coupling losses, which directly affect the overall efficiency of integrated photonic devices. Various coupling techniques have been developed to address this issue, with grating couplers and edge couplers being the most widely used solutions. Grating couplers provide a vertical coupling mechanism by redirecting incident light from a fiber into a waveguide through a diffraction-based structure. They offer key advantages such as relaxed alignment tolerances, compatibility with wafer-scale testing, and compact footprint, making them particularly attractive for high-density integrated photonic systems. In contrast, edge couplers provide higher coupling efficiency but require precisely polished facets, making fabrication and packaging more challenging. Silicon (Si) and silicon nitride (SiN) are the two most commonly used materials in photonic integration. Silicon photonics, with its high refractive index contrast, enables strong optical confinement and is CMOS-compatible, making it suitable for large-scale integration. Silicon nitride, on the other hand, offers lower propagation losses and broader spectral operation, making it ideal for nonlinear and quantum applications. Despite the extensive research on grating couplers, most studies focus on a single material. This thesis proposes optimized grating coupler designs utilizing both Si and SiN materials at 1550 nm wavelength. By optimizing design parameters such as grating period, etch depth, and coupling angle, we aim to enhance coupling efficiency while maintaining a compact footprint. The results of this study contribute to the development of high-performance integrated photonic systems by improving fiber-to-chip coupling efficiency across multiple wavelengths, thereby expanding the applicability of photonic platforms