Middle East Technical University Research Information System
Not a member yet
104897 research outputs found
Sort by
Linearly Implicit Exponential Integrators for Damped Hamiltonian Pdes
No full textWe construct second-order structure-preserving two-step linearly implicit exponential integrators for Hamiltonian partial differential equations with linear constant damping combining discrete gradient methods and polarization of the polynomial Hamiltonian function. We also construct an exponential version of the well-known one-step Kahan's method by polarizing the quadratic vector field. These integrators are applied to one-dimensional damped Burger's, Korteweg-de Vries, and nonlinear Schr & ouml;dinger equations. Preservation of the dissipation rate is demonstrated for linear and quadratic conformal invariants and for the Hamiltonians by numerical experiments
Study of the effect of ionizing γ-radiation on the structural and vibrational properties of GaS and Yb-doped GaS (GaSYb) single crystals by X-ray diffraction (XRD) and Raman spectroscopy
No full textThis study investigates the effects of gamma irradiation on the structural and vibrational properties of GaS and Yb-doped GaS (GaSYb) single crystals. X-ray diffraction (XRD) analysis reveals that irradiation induces sulfur vacancies and gallium interstitials in GaS, leading to lattice contraction, increased microstrain, and reduced crystallinity. Raman spectroscopy confirms these modifications, showing phonon frequency shifts, peak broadening, and intensity reduction due to increased phonon scattering. In contrast, Yb incorporation enhances structural stability by strengthening Ga-S bonds, reducing defect formation, and mitigating radiation-induced amorphization. These findings demonstrate that Yb doping improves the radiation resistance of GaS, making it a promising material for applications in radiation-intensive environments
Hybrid Mixed-Effect Diffusion model (H-MED) for longitudinal air quality analysis
No full textAir pollution poses severe threats to global public health and environmental sustainability, with complex spatiotemporal dynamics. Traditional air quality models face limitations in capturing longitudinal dependencies and managing computational costs at scale. To address these challenges, we propose the Hybrid MixedEffect Diffusion (H-MED) model, a novel framework integrating mixed-effects regression, Gaussian-Process Regression (GPR), Long Short-Term Memory (LSTM), and diffusion-based learning for enhanced predictive accuracy and uncertainty quantification. Applied to a comprehensive longitudinal dataset spanning 61 countries (2013-2023), H-MED demonstrated superior performance, achieving the lowest prediction errors (MAE:0.1423, RMSE:0.1925), while maintaining computational efficiency (5.69 s). The model significantly outperformed state-of-the-art approaches including advanced spatiotemporal models such as HITS, DCRNN, ST-GCN, and DeepAR by 15%-20% across all metrics. By incorporating country-level random effects and providing SHAP-based feature attribution, H-MED delivers interpretable, high-resolution forecasts that capture regional pollution heterogeneity and enable targeted policy interventions, a critical advantage over black-box deep learning models that lack explainability for environmental decision-making
LOCO Codes Can Correct as Well: Error-Correction Constrained Coding for DNA Data Storage
No full textAs a medium for cold data storage, DNA stands out as it promises significant gains in storage capacity and lifetime. However, it comes with its own data processing challenges to overcome. Constrained codes over the DNA alphabet {A, T, G, C} have been used to design DNA sequences that are free of long homopolymers to increase stability, yet effective error detection and error correction are required to achieve reliability in data retrieval. Recently, we introduced lexicographically-ordered constrained (LOCO) codes, namely DNA LOCO (D-LOCO) codes, with error detection. In this paper, we equip our D-LOCO codes with error correction for substitution errors via syndrome-like decoding, designated as residue decoding. We only use D-LOCO codewords of indices divisible by a suitable redundancy metric R(m) > 0, where m is the code length, for error correction. The idea is that the residue, which is the index modulo R(m), of the received word index equals the residue of the index error, i.e., the index difference. We find an exhaustive list of index differences due to single-substitution errors, and we are able to recover the index of the original codeword from the residue of the received word index. This requires storing a table for index errors and their residues. Having this decoding algorithm in hand, we provide the community with a construction of constrained codes forbidding runs of length higher than fixed ℓ ∈ {1,2,3} and GC-content in (Formula presented) that correct K segmented substitution errors, one per codeword. We call the proposed codes error-correction (EC) D-LOCO codes. We also give a list-decoding procedure with near-quadratic time-complexity in m to correct double-substitution errors within EC D-LOCO codewords, which has > 98.20% average success rate. The redundancy metric is projected to require 2log2(m)+O(1)-bit allocation, i.e., 2log2(m)+O(1) reduction in message bits, for a length-m codeword. Hence, our EC D-LOCO codes are projected to be capacity-approaching with respect to the error-free constrained system
Transformers in Small Object Detection: A Benchmark and Survey of State-of-the-Art
No full textTransformers have rapidly gained popularity in computer vision, especially in the field of object detection. Upon examining the outcomes of state-of-the-art object detection methods, we noticed that transformers consistently outperformed well-established CNN-based detectors in almost every video or image dataset. Small objects have been identified as one of the most challenging object types in detection frameworks due to their low visibility. This article aims to explore the performance benefits offered by such extensive networks and identify potential reasons for their Small Object Detection (SOD) superiority. We aim to investigate potential strategies that could further enhance transformers' performance in SOD. This survey presents a taxonomy of over 60 research studies on developed transformers for the task of SOD, spanning the years 2020 to 2023. These studies encompass a variety of detection applications, including SOD in generic images, aerial images, medical images, active millimeter images, underwater images, and videos. We also compile and present a list of 12 large-scale datasets suitable for SOD that were overlooked in previous studies and compare the performance of the reviewed studies using popular metrics such as mean Average Precision (mAP), Frames Per Second (FPS), and number of parameters. Researchers can keep track of newer studies on our web page, which is available at: https://github.com/arekavandi/Transformer-SOD
Strut and tie models with the dual use of the finite element method and the lattice networks
No full textStrut and tie modelling (STM) is widely recognized as a prominent design method for the design of disturbed (D) region reinforced concrete structural elements. The paramount initial step in conducting optimized reinforcement designs lies in the establishment of strut and tie geometry. In this study, a novel dual modelling approach is proposed, combining two distinct models, namely the continuum model (specifically, a finite element model) and the lattice network (comprising overlapping truss elements interconnected at every two nodes). The approach involves the application of a novel element removal algorithm based on the computed internal forces and stresses for the two lattice and finite element models working in tandem while sharing information. The iterative process continues until the convergence of the internal energy of the reduced lattice network. To evaluate the accuracy and efficacy of the proposed approach, a comprehensive demonstration is conducted employing the 20 most extensively studied design examples documented in the literature. The findings indicate that the dual system is a viable alternative method for STM design by satisfying the minimum energy requirements and without conducting laborious nonlinear finite element analysis. Furthermore, objective designs are possible from the derived STM models which can used in automated design flows
Flexural Performance of Geopolymer-Reinforced Concrete Beams Under Monotonic and Cyclic Loading: Experimental Investigation
No full textThis study investigates the flexural performance of geopolymer (zero-cement) concrete (ZCC) beams compared to normal concrete (NC) under monotonic and cyclic loading. Sixteen reinforced beams with compressive strengths of 20 and 30 Mpa and reinforcement configurations of 2Ø10 and 3Ø12 were tested to evaluate load–deflection behavior, ductility, energy absorption, and cracking characteristics. Under monotonic loading, ZCC beams achieved 9–17% higher ultimate strength and 5–30% greater mid-span deflection than NC beams, indicating superior ductility and energy dissipation. Under cyclic loading, ZCC beams demonstrated more stable hysteresis loops, slower stiffness degradation, and 8–32% higher cumulative energy absorption. ZCC specimens also sustained 8–12 cycles, corresponding to 70–90% of the monotonic displacement, whereas NC beams generally failed earlier at lower displacement levels. Increasing reinforcement ratio enhanced stiffness and load capacity but reduced deflection for both materials. Crack mapping showed finer and more uniformly distributed cracking in ZCC beams, confirming improved bond behavior between steel reinforcement and the geopolymer matrix. In addition, geopolymer concrete beams exhibited a significant enhancement in ductility, with the ductility coefficient increasing by nearly 50% compared to normal concrete under cyclic loading. Overall, the findings indicate that ZCC provides comparable or superior structural performance relative to NC, supporting its application as a sustainable, low-carbon material for flexure- and shear-critical members subjected to static and cyclic actions
MMW/THz seeker with passive ranging
No full textPassive detection and imaging systems, which operate without the need for a light source, are the subject of frequent research interest. One notable application of passive detection is passive range measurement. This study investigates passive ranging techniques, commonly utilized in infrared search and tracking (IRST) systems, within the context of the passive Millimeter Wave (MMW)/Terahertz (THz) seeker concept. Unlike conventional bands that employ mid-wave infrared (MWIR), this study proposes a rosette scanning seeker system specifically tailored for air-to-air missiles with a passive ranging capability. In addition, this study also presents the optical configuration and the operational principle of the seeker has been formalized, and relevant parameters have been identified. Additionally, the passive-ranging capability has been discussed in this seeker concept and ratio tests were carried out in indoor conditions in order to verify the passive ranging capability. Moreover, characterization of the receiver has also been proposed for MMW/THz bands. This research is expected to significantly contribute to the development of new detection and imaging systems in the MMW/THz bands in the future
