Hasil untuk "Architecture"

Xingxing Zhao, Wang Fen, Zhenwei Dai et al.

Abstract Rockfall represents a sudden and highly destructive geological hazard, posing significant risks to mountainous communities and infrastructure. This study presents an integrated dynamic risk assessment for the Jiaohua perilous rock zone in Kaizhou District, Chongqing, China, by fusing multi-source data including field investigation, UAV photogrammetry, and 3D numerical simulation. Kinematic analysis identified a critical slope angle of 57° for rockfall initiation, enabling the classification of two primary susceptibility zones. High-precision 3D simulations using RAMMS: ROCKFALL were conducted on six identified hazardous rock masses (#WY1–#WY6). The simulations delineated two distinct rockfall mechanisms: #WY1–#WY3 sources generate high-energy, short-duration impacts, achieving kinetic energies up to 1.88 × 10⁴ kJ within 10–15 s, posing a direct threat to the residential area below. Conversely, rockfalls from #WY4–#WY6 involve longer travel paths with considerable energy attenuation, yet residual kinetic energy remains capable of causing zonal damage. The simulated kinetic energies were translated into quantitative impact force estimates, forming the basis for differentiated mitigation strategies. These include active reinforcement and high-strength interception for short-range, high-energy events, and multi-level buffering with trajectory control for long-runout cases. This integrated methodology offers a scientifically grounded framework for precise hazard prevention and serves as a valuable reference for rockfall risk management in analogous geological settings, particularly within the Three Gorges Reservoir area.

Olga Krestinskaya, Mohammed E. Fouda, Ahmed Eltawil et al.

To maximize hardware efficiency and performance accuracy in Compute-In-Memory (CIM)-based neural network accelerators for Artificial Intelligence (AI) applications, co-optimizing both software and hardware design parameters is essential. Manual tuning is impractical due to the vast number of parameters and their complex interdependencies. To effectively automate the design and optimization of CIM-based neural network accelerators, hardware-aware neural architecture search (HW-NAS) techniques can be applied. This work introduces CIMNAS, a joint model-quantization-hardware optimization framework for CIM architectures. CIMNAS simultaneously searches across software parameters, quantization policies, and a broad range of hardware parameters, incorporating device-, circuit-, and architecture-level co-optimizations. CIMNAS experiments were conducted over a search space of 9.9x10^85 potential parameter combinations with the MobileNet model as a baseline and RRAM-based CIM architecture. Evaluated on the ImageNet dataset, CIMNAS achieved a reduction in energy-delay-area product (EDAP) ranging from 90.1x to 104.5x, an improvement in TOPS/W between 4.68x and 4.82x, and an enhancement in TOPS/mm^2 from 11.3x to 12.78x relative to various baselines, all while maintaining an accuracy of 73.81%. The adaptability and robustness of CIMNAS are demonstrated by extending the framework to support the SRAM-based ResNet50 architecture, achieving up to an 819.5x reduction in EDAP. Unlike other state-of-the-art methods, CIMNAS achieves EDAP-focused optimization without any accuracy loss, generating diverse software-hardware parameter combinations for high-performance CIM-based neural network designs. The source code of CIMNAS is available at https://github.com/OlgaKrestinskaya/CIMNAS.

Josh Millar, Ryan Gibb, Roy Ang et al.

Physical spaces are increasingly dense with networked devices, promising seamless coordination and ambient intelligence. Yet today, cloud-first architectures force all communication through wide-area networks regardless of physical proximity. We lack an abstraction for spatial networking: using physical spaces to create boundaries for private, robust, and low-latency communication. We introduce $\textit{Bifröst}$, a programming model that realizes spatial networking using bigraphs to express both containment and connectivity, enabling policies to be scoped by physical boundaries, devices to be named by location, the instantiation of spatial services, and the composition of spaces while maintaining local autonomy. Bifröst enables a new class of spatially-aware applications, where co-located devices communicate directly, physical barriers require explicit gateways, and local control bridges to global coordination.

Liang Qian, Xiwen Zeng, Xiaorong Liu et al.

Correlated Color Temperature (CCT) significantly influences mood, comfort, and potentially overall health. However, its impact on visitors’ visual experience in museum design remains insufficiently explored. This study aims to investigate the effects of different CCT settings (3000 K, 4500 K, 6000 K) on visual comfort within a simulated museum space. Using 3D modeling and physiological recordings, 200 participants assessed visual comfort. Consistent findings support that a CCT of 4500 K provides the highest comfort level, aligning with the observed trend in eye gaze duration. Pupil diameter variability indicates that greater comfort is associated with higher CCT values. While differences in heart rate variability (HRV) were not statistically significant, there is a tendency for HRV to increase with longer fixation durations. These findings challenge literature advocating for lower CCT values in museum lighting, emphasizing the need to balance conservation and visitor experience. This study provides empirical evidence supporting the optimization of visual comfort in museum lighting design through a CCT value of 4500 K, offering valuable insights for practitioners. However, limitations include potential scene disturbance and the simulated environment. Future studies should diversify samples and explore a broader range of CCT values.

Hun Kim, Jaewoo So

In a multi-cell network, interference management between adjacent cells is a key factor that determines the performance of the entire cellular network. In particular, in order to control inter-cell interference while providing a high data rate to users, it is very important for the base station (BS) of each cell to appropriately control the transmit power in the downlink. However, as the number of cells increases, controlling the downlink transmit power at the BS becomes increasingly difficult. In this paper, we propose a multi-agent deep reinforcement learning (MADRL)-based transmit power control scheme to maximize the sum rate in multi-cell networks. In particular, the proposed scheme incorporates a long short-term memory (LSTM) architecture into the MADRL scheme to retain state information across time slots and to use that information for subsequent action decisions, thereby improving the sum rate performance. In the proposed scheme, the agent of each BS uses only its local channel state information; consequently, it does not need to receive signal messages from adjacent agents. The simulation results show that the proposed scheme outperforms the existing MADRL scheme by reducing the amount of signal messages exchanged between links and improving the sum rate.

Yihan Wang, Ping Wang, Peng Di et al.

Panax ginseng C. A. Meyer (ginseng) is one of the most widely used traditional Chinese medicinal herbs, with its roots as the primary medicinal part garnering significant attention due to their therapeutic potential. The GRAS [GRI (Gibberellic Acid Insensitive), RGA (Repressor of GAI-3 mutant), and SCR (Scarecrow)] genes are a class of widely distributed plant-specific transcription factors that play crucial roles in various physiological processes including root formation, fruit development, hormone signaling, and stem cell maintenance. This study systematically identified 139 GRAS genes (PgGRAS) in the ginseng genome for the first time, analyzing their complexity and diversity through protein domain structure, phylogenetic relationships, gene structure, and cis-acting element prediction. Evolutionary analysis revealed that all PgGRAS members were divided into 14 evolutionary branches, including a novel species-specific subfamily PG28, with segmental duplication being the primary driver of family expansion. RNA-seq analysis uncovered tissue-specific expression patterns of the PgGRAS gene family. qRT-PCR validation demonstrated that PgGRAS48, a member of the SCL3 subfamily, was significantly highly expressed in the main root and upregulated upon GA treatment, suggesting its potential regulatory role in main root development. Therefore, this gene was selected for further investigation. Overexpression of PgGRAS48 significantly increased the main root length in Arabidopsis thaliana (A. thaliana), accompanied by elevated endogenous GA levels. Subcellular localization, molecular docking, Bimolecular Fluorescence Complementation (BIFC) and yeast two-hybrid (Y2H) experiments confirmed the interaction between PgGRAS48 (SCL3) and PgGRAS2 (DELLA) in the nucleus, revealing the molecular mechanism by which SCL3-DELLA regulates main root elongation through gibberellin (GA) biosynthesis or signaling pathways. This study elucidates the molecular network of the GRAS family in root development in ginseng, providing key targets for the targeted improvement of root architecture in medicinal plants.

Jie Peng, Yuxi Zou, Hao Zhang et al.

Globally, recycled building materials have attracted much attention, but the ambiguity of the use of recycled building materials makes it difficult for the building material remanufacturer (BMR) to compete with the building material manufacturer (BMM). Brand building is an important strategic tool for enterprises to increase product competitiveness. From the new perspective of the supply chain, this paper aims to examine the decision-making behavior of enterprises under two scenarios of consumer ambiguity neutrality and ambiguity tolerance and to analyze the impact of ambiguity tolerance on the pricing decisions of building materials supply chains in a brand-building scenario. This paper constructs a building material supply chain game model consisting of the BMM and BMR, according to the cognitive–affective personality system (CAPS) theory and through the Stackelberg game. The main findings are as follows. (1) Strengthening brand building can mitigate the negative impact of ambiguity tolerance on new product pricing. The selling price of recycled building materials is positively related to ambiguity tolerance. (2) When the BMM has higher brand value, there is a U-shaped trend between profit and ambiguity tolerance at a cost coefficient above the threshold value of 0.61. (3) When the BMR has higher brand value, profit is negatively related to ambiguity tolerance at operational inefficiencies and cost coefficients below the threshold value of 0.45. Otherwise, profits and ambiguity tolerance follow a U-shaped trend. This paper not only expands the research on brand building and ambiguity tolerance but also provides theoretical guidance for enterprises to make effective decisions in response to consumers’ ambiguity psychology.

Mika Markus Müller, Alexander Richard Manfred Borst, Konstantin Lübeck et al.

Artificial Intelligence (AI) has witnessed remarkable growth, particularly through the proliferation of Deep Neural Networks (DNNs). These powerful models drive technological advancements across various domains. However, to harness their potential in real-world applications, specialized hardware accelerators are essential. This demand has sparked a market for parameterizable AI hardware accelerators offered by different vendors. Manufacturers of AI-integrated products face a critical challenge: selecting an accelerator that aligns with their product's performance requirements. The decision involves choosing the right hardware and configuring a suitable set of parameters. However, comparing different accelerator design alternatives remains a complex task. Often, engineers rely on data sheets, spreadsheet calculations, or slow black-box simulators, which only offer a coarse understanding of the performance characteristics. The Abstract Computer Architecture Description Language (ACADL) is a concise formalization of computer architecture block diagrams, which helps to communicate computer architecture on different abstraction levels and allows for inferring performance characteristics. In this paper, we demonstrate how to use the ACADL to model AI hardware accelerators, use their ACADL description to map DNNs onto them, and explain the timing simulation semantics to gather performance results.

Rodrigo Moreira, Rodolfo S. Villaca, Moises R. N. Ribeiro et al.

Network Slicing (NS) has transformed the landscape of resource sharing in networks, offering flexibility to support services and applications with highly variable requirements in areas such as the next-generation 5G/6G mobile networks (NGMN), vehicular networks, industrial Internet of Things (IoT), and verticals. Although significant research and experimentation have driven the development of network slicing, existing architectures often fall short in intrinsic architectural intelligent security capabilities. This paper proposes an architecture-intelligent security mechanism to improve the NS solutions. We idealized a security-native architecture that deploys intelligent microservices as federated agents based on machine learning, providing intra-slice and architectural operation security for the Slicing Future Internet Infrastructures (SFI2) reference architecture. It is noteworthy that federated learning approaches match the highly distributed modern microservice-based architectures, thus providing a unifying and scalable design choice for NS platforms addressing both service and security. Using ML-Agents and Security Agents, our approach identified Distributed Denial-of-Service (DDoS) and intrusion attacks within the slice using generic and non-intrusive telemetry records, achieving an average accuracy of approximately $95.60\%$ in the network slicing architecture and $99.99\%$ for the deployed slice -- intra-slice. This result demonstrates the potential for leveraging architectural operational security and introduces a promising new research direction for network slicing architectures.

Cristobal Ortega, Yann Falevoz, Renaud Ayrignac

Large Language Models (LLMs) have become essential in a variety of applications due to their advanced language understanding and generation capabilities. However, their computational and memory requirements pose significant challenges to traditional hardware architectures. Processing-in-Memory (PIM), which integrates computational units directly into memory chips, offers several advantages for LLM inference, including reduced data transfer bottlenecks and improved power efficiency. This paper introduces PIM-AI, a novel DDR5/LPDDR5 PIM architecture designed for LLM inference without modifying the memory controller or DDR/LPDDR memory PHY. We have developed a simulator to evaluate the performance of PIM-AI in various scenarios and demonstrate its significant advantages over conventional architectures. In cloud-based scenarios, PIM-AI reduces the 3-year TCO per queries-per-second by up to 6.94x compared to state-of-the-art GPUs, depending on the LLM model used. In mobile scenarios, PIM-AI achieves a 10- to 20-fold reduction in energy per token compared to state-of-the-art mobile SoCs, resulting in 25 to 45~\% more queries per second and 6.9x to 13.4x less energy per query, extending battery life and enabling more inferences per charge. These results highlight PIM-AI's potential to revolutionize LLM deployments, making them more efficient, scalable, and sustainable.

Haoyi Zhang, Jiahao Song, Xiaohan Gao et al.

Analog Computing-in-Memory (ACIM) is an emerging architecture to perform efficient AI edge computing. However, current ACIM designs usually have unscalable topology and still heavily rely on manual efforts. These drawbacks limit the ACIM application scenarios and lead to an undesired time-to-market. This work proposes an end-to-end automated ACIM based on a synthesizable architecture (EasyACIM). With a given array size and customized cell library, EasyACIM can generate layouts for ACIMs with various design specifications end-to-end automatically. Leveraging the multi-objective genetic algorithm (MOGA)-based design space explorer, EasyACIM can obtain high-quality ACIM solutions based on the proposed synthesizable architecture, targeting versatile application scenarios. The ACIM solutions given by EasyACIM have a wide design space and competitive performance compared to the state-of-the-art (SOTA) ACIMs.

Danilo Fernandes, Douglas L. L. Moura, Gean Santos et al.

The rapid urbanization growth has underscored the need for innovative solutions to enhance transportation efficiency and safety. Intelligent Transportation Systems (ITS) have emerged as a promising solution in this context. However, analyzing and processing the massive and intricate data generated by ITS presents significant challenges for traditional data processing systems. This work proposes an Edge-based Data Lake Architecture to integrate and analyze the complex data from ITS efficiently. The architecture offers scalability, fault tolerance, and performance, improving decision-making and enhancing innovative services for a more intelligent transportation ecosystem. We demonstrate the effectiveness of the architecture through an analysis of three different use cases: (i) Vehicular Sensor Network, (ii) Mobile Network, and (iii) Driver Identification applications.

Ramakrishna Kanungo, Swamynathan Siva, Nathaniel Bleier et al.

Mitigating losses from supply and demand volatility in the semiconductor supply chain and market has traditionally been cast as a logistics and forecasting problem. We investigate how the architecture of a family of chips influences how it is affected by supply and demand uncertainties. We observe that semiconductor supply chains become fragile, in part, due to single demand paths, where one chip can satisfy only one demand. Chip architects can enable multiple paths to satisfy a chip demand, which improves supply chain resilience. Based on this observation, we study composition and adaptation as architectural strategies to improve resilience to volatility and also introduce a third strategy of dispersion. These strategies allow multiple paths to satisfy a given chip demand. We develop a model to analyze the impact of these architectural techniques on supply chain costs under different regimes of uncertainties and evaluate what happens when they are combined. We present several interesting and even counterintuitive observations about the product configurations and market conditions where these interventions are impactful and where they are not. In all, we show that product redesign supported by architectural changes can mitigate nearly half of the losses caused by supply and demand volatility. As far as we know, this is the first such investigation concerning chip architecture.

S. Anush Lakshman, S. Akash, J. Cynthia et al.

Number of IoT enabled devices are being tried and introduced every year and there is a healthy competition among researched and businesses to capitalize the space created by IoT, as these devices have a great market potential. Depending on the type of task involved and sensitive nature of data that the device handles, various IoT architectures, communication protocols and components are chosen and their performance is evaluated. This paper reviews such IoT enabled devices based on their architecture, communication protocols and functions in few key socially relevant fields like health care, farming, firefighting, women/individual safety/call for help/harm alert, home surveillance and mapping as these fields involve majority of the general public. It can be seen, to one's amazement, that already significant number of devices are being reported on these fields and their performance is promising. This paper also outlines the challenges involved in each of these fields that require solutions to make these devices reliable

Valerie Fraser

Ali Hassan Abdellah Mohamed Eldaly

The Art of Animation is always moving towards creativity and distinction, and as long as animation artists succeed in drowning joy and happiness on the faces of adults and children of the audience of that beautiful art through different types and works of art with its multiple techniques whether two - dimensional or three - dimensional animation technology, clay ,Cut out techniques or moving the puppet technique of moving frame by frame it's production forms vary between long and narrative films, short films, serials, promotional advertisements, or animation breaks displayed on satellite channels , we here in this research to shed light on a short animation separator shown on satellite channel during the holy month of Ramadan. Animation it is a Tow - dimensional animation technique to implement rhythmic interval animation. The research also addressed in this rhythmic interval to shed light on the growth of the rhythmic and creativity role of tow - dimensional movement through a rhythmic musical piece that was played with strings, which is a song, and vocalist. The Egyptian . who is full of many traditional songs that have been stuch in the minds and hearts of the Egyptian people old and young .That song was composed and composed by ( Ahmed Abd El-Qader from 1916th aged ), The Egyptian artist and singer who accompanied the opening of the Egyptian Radio at 1934th year and was one of the first singers who participated in singing in its programs since the first week.The rhythmic musician has a role in highlighting and intensifying the dramatic event in animation films, so there must be a dynamic connection between what appears with in the frame of the picture and what the recipients hear, as the music that is not synchronized with the movement may lead to a dramatic negative result in the breaks or animated films. From here this must be translated. The inter connectedness of rhythmic movement and music makes the animation to the dramatic through the perception of the animator.

N. Aslan, D. Koc-San

The aims of this study are to detect the land-cover maps and land surface temperatures using Landsat time series and analyse the relation between the land-cover and land surface temperatures (LST) and their changes in time. For these purposes initially, land-cover maps were generated rapidly using land cover indices and automatic thresholding. The land-cover indices used in this study are Normalized Difference Vegetation Index (NDVI), Modified Normalized Difference Water Index (MNDWI), Index-Based Built-Up Index (IBI), Modified Bare-Soil Index (MBI), Plastic-Mulched Landcover Index (PMLI), Plastic Greenhouse Index (PGI) and Normalized Burn Ratio Thermal (NBRT) Index. Then, using the thermal bands of Landsat satellites, LST maps were created. Finally, the land-cover and LST changes were examined. The Kumluca district of Antalya, which includes extensive greenhouse areas as well as urban, vegetation, bareland, water, was selected as study area. Between the years 2004 and 2021, within the study area the greenhouse areas increased significantly, the urban area expanded and some areas exposed to fire, especially the fire in 2016. Therefore, the images within this time period were used. The overall accuracies for land-cover maps were computed as 76%, 79%, 79%, 89% and 86% for the years 2004, 2009, 2013, 2017 and 2021, respectively. The results obtained from the study reveal that while greenhouse and urban areas were increased, the vegetation areas were decreased significantly within this time period. In addition, generally increases were observed for LST values of all land-cover classes and the highest LST values were detected for the burned, bareland, urban and greenhouse areas.

Marco Antonio Zanata Alves, Sairo Santos, Aline S. Cordeiro et al.

Data movement is one of the main challenges of contemporary system architectures. Near-Data Processing (NDP) mitigates this issue by moving computation closer to the memory, avoiding excessive data movement. Our proposal, Vector-In-Memory Architecture(VIMA), executes large vector instructions near 3D-stacked memories using vector functional units and uses a small data cache to enable short-term data reuse. It provides an easy programming interface and guarantees precise exceptions. When executing stream-behaved applications using a single core, VIMA offers a speedup of up to 26x over a CPU system baseline with vector operations in a single-core processor while spending 93% less energy.

Dongseok Im, Gwangtae Park, Zhiyong Li et al.

As the number of deep neural networks (DNNs) to be executed on a mobile system-on-chip (SoC) increases, the mobile SoC suffers from the real-time DNN acceleration within its limited hardware resources and power budget. Although the previous mobile neural processing units (NPUs) take advantage of low-bit computing and exploitation of the sparsity, it is incapable of accelerating high-precision and dense DNNs. This paper proposes energy-efficient signed bit-slice architecture which accelerates both high-precision and dense DNNs by exploiting a large number of zero values of signed bit-slices. Proposed signed bit-slice representation (SBR) changes signed $1111_{2}$ bit-slice to $0000_{2}$ by borrowing a $1$ value from its lower order of bit-slice. As a result, it generates a large number of zero bit-slices even in dense DNNs. Moreover, it balances the positive and negative values of 2's complement data, allowing bit-slice based output speculation which pre-computes high order of bit-slices and skips the remaining dense low order of bit-slices. The signed bit-slice architecture compresses and skips the zero input signed bit-slices, and the zero skipping unit also supports the output skipping by masking the speculated inputs as zero. Additionally, the heterogeneous network-on-chip (NoC) benefits the exploitation of data reusability and reduction of transmission bandwidth. The paper introduces a specialized instruction set architecture (ISA) and a hierarchical instruction decoder for the control of the signed bit-slice architecture. Finally, the signed bit-slice architecture outperforms the previous bit-slice accelerator, Bit-fusion, over $\times3.65$ higher area-efficiency, $\times3.88$ higher energy-efficiency, and $\times5.35$ higher throughput.

Inés Magdalena Campos-García Calderón, Doraliza Olivera Mendoza

La Unidad Vecinal n.° 3 (UV3) fue un planteamiento de vivienda social basado en la teoría de la Neighborhood-Unit y la Ciudad-Satélite, donde los Espacios Libres Planificados (ELP) fueron relevantes para la salubridad y el desarrollo comunitario de la población, por lo que fueron ocupados y sus características físico-arquitectónicas fueron transformadas; el objetivo de esta investigación fue identificar la transformación por apropiación de los ELP en la UV3. Mediante un enfoque cualitativo se llevó a cabo un análisis comparativo gráfico del planteamiento original versus la situación actual, análisis documental y observación de campo. Se encontraron cambios a partir de la ubicación de elementos materiales para delimitar y subdividir espacios y la inclusión de estos al espacio residencial, los cambios de uso de área verde colectiva a espacio individual de la vivienda contigua, y la colocación de elementos simbólicos de reconocimiento. Lo cual resulta de distintos tipos de apropiación: según el agente, la naturaleza y las consecuencias. Se concluye que la transformación de los ELP del planteamiento original ha sido posible por la desmesura de sus áreas y fue resultado de diferentes formas de apropiación que han generado un perfil urbano informal.