5916 research outputs found
Sort by
Differential privacy preserving based framework using blockchain for internet-of-things
No full textThe Internet of Things (IoT) has enabled the collection of vast amounts of data that can be used to improve various aspects of our lives. However, the astronomical volume of data generated by these IoT devices has raised significant concerns pertaining to privacy preservation. The amalgamation of the Internet of Things (IoT) with blockchain technology has engendered a promising solution for securing and managing IoT data, but it is still susceptible to privacy breaches. Recently, differential privacy (DP) has been proposed as a promising technique to alleviate these issues. In this paper, we design and propound a complete end-to-end blockchain-based architecture by implementing differential privacy at the stream level generated by IoT devices by deploying Laplace noise and Gaussian noise utilizing low complex cryptography mechanism and fast convergence consensus protocol to surmount the privacy preservation issues in IoT based blockchain network. Our novel DP-based framework introduces the concept of privacy levels as low, medium, and high as set by the data owner and also analyzes the impact of different parameters on the effectiveness of the approach and provides recommendations for tuning them. The workflow of our proposed framework consists of three phases: Data generation phase, Data Sharing phase, and Data Analysis phase. During the Data generation phase, the data owner will first determine the desired level of privacy protection (low, medium, high) and set the privacy budget (epsilon) and sensitivity (delta) of the data. Based on the budget value, the privacy module will generate noise from either Laplace or Gaussian distribution as requested by the data owner. The Data Sharing phase is mainly responsible for transmitting and processing the transactions inside the blockchain network. This is followed by the data analysis phase, which will check for the budget value and the amount of noise added to the data before the noisy data is handed over to the end user. We demonstrate the efficacy of our approach through multiple experimental evaluations and simulation results evince that our approach attains high levels of privacy preservation while upholding data utility and blockchain consistency. Overall, our proposed framework provides a promising solution to the privacy challenges in IoT-based blockchain systems, offering adjustable privacy levels to accommodate different privacy requirements. This DP-based approach and the adjustable privacy levels ensure alignment with the growing regulatory requirements for data privacy, such as GDPR, demonstrating compliance with these regulations and building trust with customers. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024
JugglePAC: A pipelined accumulation circuit
No full textReducing a set of numbers to a single value is a fundamental operation in applications such as signal processing, data compression, scientific computing, and neural networks. Accumulation, which involves summing a dataset to obtain a single result, is crucial for these tasks. Due to hardware constraints, large vectors or matrices often cannot be fully stored in memory and must be read sequentially, one item per clock cycle. For high-speed inputs, such as rapidly arriving floating-point numbers, pipelined adders are necessary to maintain performance. However, pipelining introduces multiple intermediate sums and requires delays between back-to-back datasets unless their processing is overlapped. In this paper, we present JugglePAC, a novel accumulation circuit designed to address these challenges. JugglePAC operates quickly, is area-efficient, and features a fully pipelined design. It effectively manages back-to-back variable-length datasets while consistently producing results in the correct input order. Compared to the state-of-the-art, JugglePAC achieves higher throughput and reduces area complexity, offering significant improvements in performance and efficiency
Mask-to-height: A YOLOv11-based architecture for joint building instance segmentation and height classification from satellite imagery
No full textAccurate building instance segmentation and height classification are critical for urban planning, 3D city modeling, and infrastructure monitoring. This paper presents a detailed analysis of YOLOv11, the recent advancement in the YOLO series of deep learning models, focusing on its application to joint building extraction and discrete height classification from satellite imagery. YOLOv11 builds on the strengths of earlier YOLO models by introducing a more efficient architecture that better combines features at different scales, improves object localization accuracy, and enhances performance in complex urban scenes. Using the DFC2023 Track 2 dataset - which includes over 125,000 annotated buildings across 12 cities - we evaluate YOLOv11's performance using metrics such as precision, recall, F1 score, and mean average precision (mAP). Our findings demonstrate that YOLOv11 achieves strong instance segmentation performance with 60.4% mAP @ 50 and 38.3% mAP @ 50-95 while maintaining robust classification accuracy across five predefined height tiers. The model excels in handling occlusions, complex building shapes, and class imbalance, particularly for rare high-rise structures. Comparative analysis confirms that YOLOv11 outperforms earlier multitask frameworks in both detection accuracy and inference speed, making it well-suited for real-time, large-scale urban mapping. This research highlights YOLOv11's potential to advance semantic urban reconstruction through streamlined categorical height modeling, offering actionable insights for future developments in remote sensing and geospatial intelligence. © 2025 IEEE.TÜBİTA
Mavi ışık yayan diyotların ve beyaz ışık dönüşümünün termal ve optik özelliklerine dair bir inceleme.
No full textThe evolution of lighting technologies has progressed from primitive fire-based sources to advanced solid-state devices, with Light-Emitting Diodes (LEDs) emerging as a cornerstone of modern illumination due to their energy efficiency and versatility. However, the performance and longevity of LEDs are critically dependent on effective thermal management, as excessive heat generation can degrade efficiency and lead to premature failure. This work examines natural convection heat transfer from millimetric surfaces using LED chip packages as a representative system. Experiments were conducted at electrical currents between 40–120 mA, measuring junction and fluid temperatures, and results were validated against CFD simulations with good agreement. Analysis revealed that conventional Nusselt number correlations deviate significantly at this scale, particularly for horizontal and vertical chip surfaces. This study examines the thermal–optical behavior of phosphor-converted LEDs using experimental testing and CFD modeling. LED packages with phosphor concentrations of 0.99 wt.%, 1.3 wt.%, and 2.98 wt.% were fabricated and tested under controlled conditions. Junction temperatures and chromaticity coordinates were measured, while a validated CFD model reproduced the results. At 0.99 wt.%, the CIE coordinates were approximately (0.26, 0.21). When the phosphor loading was increased to 1.3 wt.%, the CIE coordinates shifted to about (0.3, 0.26). At the highest concentration of 2.98 wt.%, the junction temperature rose by ~1.7 °C, and the CIE coordinates changed further to around (0.43, 0.44). The agreement between experiments and simulations confirms the accuracy of the modeling approach and shows the direct relationship between phosphor loading, junction heating, and chromaticity shift. The findings underscore the importance of optimizing thermal management strategies for LEDs, particularly in high-power applications. The study contributes to the advancement of LED technology by providing insights into heat dissipation mechanisms and proposing practical solutions for enhancing optical and thermal performance for the added benefits of reliability.Aydınlatma teknolojilerinin evrimi, ilkel ateş tabanlı kaynaklardan gelişmiş katı hâl aygıtlarına kadar ilerlemiş olup, enerji verimliliği ve çok yönlülüğü sayesinde Işık Yayan Diyotlar (LED’ler) modern aydınlatmanın temel taşlarından biri hâline gelmiştir. Bununla birlikte, LED’lerin performansı ve ömrü, etkin bir ısıl yönetim uygulamasına kritik düzeyde bağlıdır; zira aşırı ısı üretimi, verimliliğin azalmasına ve erken arızalara yol açabilmektedir. Bu çalışma, LED çip paketleri temsilî bir sistem olarak kullanılarak milimetrik yüzeylerden doğal taşınımla gerçekleşen ısı transferini incelemektedir. Deneyler, 40–120 mA arası elektrik akımlarında gerçekleştirilmiş, eklem (junction) ve akışkan sıcaklıkları ölçülmüş ve sonuçlar Hesaplamalı Akışkanlar Dinamiği (HAD) benzetimleri ile doğrulanmış, iyi bir uyum gözlenmiştir. Analizler, geleneksel Nusselt sayısı korelasyonlarının bu ölçekte, özellikle yatay ve dikey çip yüzeylerinde önemli sapmalar gösterdiğini ortaya koymuştur. Bu çalışma ayrıca, deneysel testler ve HAD modellemesi kullanılarak fosfor dönüştürmeli LED’lerin ısıl-optik davranışını incelemektedir. Ağırlıkça %0.99, %1.3 ve %2.98 fosfor içeriğine sahip LED paketleri üretilmiş ve kontrollü koşullarda test edilmiştir. Eklem sıcaklıkları ve renk koordinatları ölçülmüş, doğrulanmış HAD modeli sonuçları yeniden üretmiştir. %0.99 fosfor yüklemesinde CIE koordinatları yaklaşık (0.27, 0.22) bulunmuştur. Fosfor yüklemesi %1.3’e çıkarıldığında koordinatlar yaklaşık (0.30, 0.26)’e kaymıştır. En yüksek yoğunluk olan %2.98’de eklem sıcaklığı yaklaşık 1.7 °C artmış, CIE koordinatları ise yaklaşık (0.43, 0.44)’e değişmiştir. Deneyler ile simülasyonlar arasındaki uyum, modelleme yaklaşımının doğruluğunu teyit etmekte ve fosfor yüklemesi, eklem ısınması ve renk kayması arasındaki doğrudan ilişkiyi göstermektedir. Elde edilen bulgular, özellikle yüksek güçlü uygulamalarda LED’ler için ısıl yönetim stratejilerinin optimize edilmesinin önemini vurgulamaktadır. Bu çalışma, ısı dağılım mekanizmalarına ilişkin kavrayış sağlayarak ve optik ile ısıl performansı artırmaya yönelik pratik çözümler önererek LED teknolojisinin gelişimine katkıda bulunmaktadır
Augmentation of pool boiling heat transfer with open micro-channel surfaces
No full textNucleate pool boiling is an efficient heat transfer mechanism widely utilized in thermal management applications such as electronic cooling and nuclear power systems. Enhancing the heat transfer coefficient (HTC) and the critical heat flux (CHF) is crucial for improving system performance and preventing thermal failures. Micro-channel structures have proven effective in enhancing nucleate boiling by modifying fluid dynamics and increasing surface area for heat dissipation. This study examines the effect of micro-channel integration on nucleate pool boiling performance. Circular copper heated surfaces with micro-channels of varying depths (0.5–1.5 mm) having the pitch and width were kept constant at 0.5 mm. The surface area factor (SAF) varied between 1.64 and 3.87. Experiments were conducted in deionized water (DI) under saturation temperature at 1 atm. Results indicate a notable 421% and %848 enhancement in CHF and HTC compared to baseline plain surfaces respectively, demonstrating the effectiveness of micro-channel design in improving heat dissipation. Additionally, CHF performance was evaluated, highlighting the role of structured surfaces in thermal management. These findings contribute to the optimization of micro-channel configurations for advanced heat transfer applications.Özyeğin Üniversites
Improvement of over-the-air antenna performance tests for electric vehicle alternating current chargers
No full textThis paper aims to provide robust cellular connectivity for alternating current electric vehicle (AC EV) chargers by implementing reliable cellular communication protocols and optimizing antenna performance. Key performance metrics such as Total Radiated Power (TRP) and Total Isotropic Sensitivity (TIS) were used to scale and assess connectivity between the LTE module, antennas, and base station. A root-cause analysis identified factors causing poor antenna performance, and theoretical data supported solutions to streamline testing processes. The results offer a framework to enhance connectivity, ensuring remote control of AC EV chargers as mandated by the Open Charge Point Protocol (OCPP)
Drowsiness and fatigue recognition systems for connected vehicles, 6G and the EU AI Act
No full textThe European Union (EU) has passed binding rules mandating the integration of drowsiness and fatigue detection systems in vehicles. This paper explores the impact of future 6G standards in advancing such in-vehicle monitoring systems. The paper, from an interdisciplinary legal and technical perspective, specifically looks at the requirements for AI-based drowsiness and fatigue recognition that have been established by the EU AI Act (Regulation (EU) 2024/1689) passed in 2024. Moreover, it discusses the secure technical implementation of those requirements as well as societal and ethical challenges and impacts on that context.EU-COST Associatio
Solid state joining of stainless steel and aluminum via interlayer friction stir spot welding
No full textInterlayer friction stir spot welding represents an advanced solid state joining technology, developed to manufacture both similar and dissimilar alloys. The incorporation of an interlayer material plays a pivotal role in preventing keyhole formation at the weld interface. This study utilizes experimental design to explore the joining of 304 stainless steel and 6061 aluminum alloy. Three key process parameters including tool rotation speed, plunge depth, and interlayer diameter are evaluated at three distinct levels. The findings indicate that the highest lap shear force of 3.66 kN can be attained by employing the optimized processing conditions. Microstructural studies reveal that the interlayer material is effectively integrated with the lower sheet, facilitated by the heat generated during the frictional process, resulting in a dendritic microstructure within the nugget zone. Moreover, fracture surface examinations demonstrate transgranular failure, providing insights on the mechanical behavior of the joints.BAGEP Award of the Science Academy ; Özyeğin Universit
Generative design research for a culturally sensitive subject: Exploring menstrual practices and product experiences to inspire design
No full textThe intimate nature of menstruation and the taboos surrounding it cause numerous challenges for conducting research in this field. To explore menstrual practices and product experiences, in this study, a generative workshop study was designed with participatory research tools and co-design exercises, following a semi-structured interview schedule. The generative research was seen as having the potential to overcome the intimate nature of the research subject and to gather design insights for menstruation. The workshop sessions include physical and visual research tools and generative exercises that are deliberately thought to remind the participants of the context of use, evoke dialog, and inspire new ideas. Understanding the underlying complexities and motives behind menstruation is crucial for the development of better solutions, products, and/or services. The data obtained from the workshops were analyzed to reveal design criteria for menstruation products and experiences by following the procedures of grounded theory. "Failures," "limitations," "social codes," "tactics," and "improvements" have been identified and discussed as design criteria that could inspire designers
Effect of PCE anionic charge density on fly ash cementitious system-PCE compatibility
No full textIn this study, the compatibility of polycarboxylate-based water-reducing admixtures (PCEs) with cementitious systems containing fly ash (FA) was investigated. For this purpose, PCEs with carboxylate, phosphate, and sulfonate anionic groups having different anionic charge densities were synthesized. The effects of PCEs on fresh properties and compressive strength of cementitious systems containing FA were investigated. The PCE with 9% phosphate substitution and high anionic charge density was found to be the most effective, requiring the least amount for the target flow. Similarly, in terms of the PCE requirement for the minimum Marsh funnel flow time and rheological parameters, the best performance was obtained with 5% sulfonate substituted PCE having high anionic charge density. While FA had a positive effect on the PCE requirement and consistency retention of the mixtures; it had a negative effect on Marsh funnel flow time, rheological properties, and compressive strength. However, the rheological properties of the mortar mixtures were not adversely affected by the FA substitution as much as that of the paste mixtures. Regarding the 28-day compressive strength of mortar mixtures, the optimum FA substitution ratio was 15%. Fly ash substitution above this level reduced the compressive strength at all ages including 28-day strength. Anionic charge density variation of PCE had no significant influence on the compressive strength of the mortars.TÜBİTA