Architectural Design Space Research Articles

CloudSatNet-1: FPGA-Based Hardware-Accelerated Quantized CNN for Satellite On-Board Cloud Coverage Classification

CubeSats, the nanosatellites and microsatellites with a wet mass up to 60 kg, accompanied by the cost decrease of accessing the space, amplified the rapid development of the Earth Observation industry. Acquired image data serve as an essential source of information in various disciplines like environmental protection, geosciences, or the military. As the quantity of remote sensing data grows, the bandwidth resources for the data transmission (downlink) are exhausted. Therefore, new techniques that reduce the downlink utilization of the satellites must be investigated and developed. For that reason, we are presenting CloudSatNet-1: an FPGA-based hardware-accelerated quantized convolutional neural network (CNN) for satellite on-board cloud coverage classification. We aim to explore the effects of the quantization process on the proposed CNN architecture. Additionally, the performance of cloud coverage classification by biomes diversity is investigated, and the hardware architecture design space is explored to identify the optimal FPGA resource utilization. Results of this study showed that the weights and activations quantization adds a minor effect on the model performance. Nevertheless, the memory footprint reduction allows the model deployment on low-cost FPGA Xilinx Zynq-7020. Using the RGB bands only, up to 90% of accuracy was achieved, and when omitting the tiles with snow and ice, the performance increased up to 94.4% of accuracy with a low false-positive rate of 2.23% for the 4-bit width model. With the maximum parallelization settings, the hardware accelerator achieved 15 FPS with 2.5 W of average power consumption (0.2 W increase over the idle state).

From System Architecting to System Design and Optimization: A Link Between MBSE and MDAO

AbstractOptimization of system architectures can help deal with finding better system architectures in a large design space plagued by combinatorial explosion of alternatives. To enable architecture optimization, the design space should therefore be formalized into a numerical optimization problem, and it should be possible to quantitatively evaluate architecture alternatives. This paper presents a methodology for generating and modeling architecture design spaces using the Architecture Design Space Graph (ADSG), and using collaborative Multidisciplinary Design Analysis and Optimization (MDAO) techniques to evaluate architectures. Collaborative MDAO leverages disciplinary expertise while ensuring that analysis tools exchange data consistently and correctly using a central data schema. The problem solved in this paper is the missing link between architecture optimization and collaborative MDAO: the reflection of generated architectures in the central data schema. It is solved by the authors by mapping architecture components and Quantities of Interest (QOIs) to the central data schema using Data Schema Operations (DSOs). Such a mapping also assists the user in identifying missing or unnecessary disciplinary analysis tools. Three web‐based software tools implementing the methodology are presented. Finally, the methodology and tools are demonstrated using the design of a supersonic business jet as an example.

Performance of superconducting quantum computing chips under different architecture designs

Existing and near-term quantum computers can only perform two-qubit gates between physically connected qubits. Research has been done on compilers to rewrite quantum programs to match hardware constraints. However, the quantum processor architecture, in particular the qubit connectivity and topology, still lacks enough discussion, while it potentially has a huge impact on the performance of the quantum algorithms. We perform a quantitative and comprehensive study on the quantum processor performance under different qubit connectivity and topology. We select ten representative design models with different connectivities and topologies from quantum architecture design space and benchmark their performance by running a set of standard quantum algorithms. It is shown that a high-performance architecture almost always comes with a design with a large connectivity, while the topology shows a weak influence on the performance in our experiment. Different quantum algorithms show different dependence on quantum chip connectivity and topologies. This work provides quantum computing researchers with a systematic approach to evaluating their processor design.

Aesthetic aspect of architectural space development for creative activity of children in a modern school

Purpose: The aim of this work is to determine the architectural space components for the creative activity of children in a modern school. Design/methodology: The description of modern design and construction of school buildings in Russia and abroad; identification of factors influencing the formation of the aesthetic aspect of the architectural space of a modern school; theoretical modeling of typological layout of individual structural elements of the architectural space for creative activities in a modern school; analysis of the functional zoning and compositional parameters of the architectural space to identify the architectural space qualitative parameters disclosing the creative potential of students. Research findings: Identification of the main algorithms of the architectural space design for the creative activity of students in a school with a specific atmosphere that has a beneficial effect on the perception of new knowledge and disclosure of the creative potential of students. Practical implications: The obtained results and proposed recommendations can be used in green architecture and development of regulatory documents on the school building design; the program preparation in project activities; in studying the individual structural components of the architectural environment within the artistic and aesthetic aspect. Originality/value: Architectural means determine the formation of the aesthetically oriented architectural environment for creative activities in modern schools and identify the qualitative parameters of the given space. The architectural environment of the creative activity represents a new type of individual structural elements of a modern school building focused on a disclosure and growth of creative potential of students with optimum opportunities.

Akvatektura parka - prikaz metodološkog ogleda u procesu projektovanja linijskog parka u Beogradu

The aim of this paper is to interpret the term Aquatecture in the design process for Zone 8 of Belgrade Linear Park, as a case study for understanding the concept and methodology used, as well as their application in operational architectural and urban practice. The paper describes the methodological procedure applied to structuring the space using water, with the aim of designing ambiences inspired by nature. This methodological experiment is illustrated with graphical representations of the process itself, in which the colors and shapes of water are analyzed through the observation of water droplets under the microscope. The research hypothesis is that one drop of water contains information important for architectural activities, and therefore the research offers a set of procedures and methods that can be applied and controlled through the creative act of coloring and shaping, or more precisely the creation of Spatial Images. The methodological procedure in the research can be described as poetic and artistic, at the same time containing a scientific element in the analysis and observation of the water sample under the microscope. All other visual perceptual observations and results in the interpretation of the observed environment and the water droplets, as well as their colors and shapes, are offered as extremely intuitive. The research is presented in the form of a Tableau consisting of seven parts and a set of photographs, drawings and images, as well as tone maps (color palette), according to which the results are applied through Spatial Images. Spatial images are interpreted as Ambience inspired by nature, at a specific location: Zone 8 of Belgrade Linear Park, as a case study for researching this methodological procedure. The data obtained in this way is applied in the structuring of the park's aquatecture and pavilions. The waterflows in the park itself are explained in the text, as well as the spatial structures such as the landscape fountain on the square and the main park pavilion itself, which is interpreted as a one-of-a-kind House of water. Therefore, as a product of the work, an environmental project inspired by nature is offered as the Aquatecture of Zone 8 in Belgrade Linear Park, illustrating the practical application of water as a methodological tool, and an inspiration in the processes of architectural design and water space structuring. In this way, the case study of the Conceptual Architectural and Urban Design of Zone 8 of the Linear Park in Belgrade offers new interpretations and understandings of Aquatecture in Belgrade in professional practice. More precisely, it re-examines ways of applying and understanding the concepts of the Blue Green Dream (BGD) and Nature Based Solutions (NBS) in shaping public spaces in the city.

Strategies for the Architectural Space Design of Special Education Resource Center in China

In view of the current problems in China’s special education resource centers, such as lack of functional cognition and unreasonable architectural function and spatial organization, which result in the inability to coordinate regional education resources, the formulation and implementation of special education ISP program are undertaken, and the development of integrated education is promoted. This paper draws on the concept of integrated education in developed areas and the design wisdom of resource centers. The research is carried out from the perspective of meeting the behavior patterns and special needs of different users, and the perfect and reasonable functional composition and the key points of special space design are put forward, and the feasible design strategies are expressed through establishing the model and plane layout.

On Transforming Unused Urban Spaces to Social Housing for Self-Employed People in Ho Chi Minh City: An Architectural Space Design Proposal

As with many metropolitan areas, social housing (SH) provision, which can improve living standards and social welfare, is crucial for urban socio-economic development strategies in Ho Chi Minh City (HCMC). However, there have been issues relating to promoting social housing in the Vietnamese context resulting from the property market and the design. The former is a failure to attract investors. The latter relates to lacking housing models for the mid- and low-income communities. Currently, many low-income families who have low access to the general job market in HCMC have to make a living by running their own business at home. This situation leads to low-income housing establishments in some residential areas. Thus, the planning approach in social housing needs to solve both the demand for low-cost housing and promoting self-employed activities. In this paper, mixed methods, including observation, questionnaires, interviews, data aggregation, and comparison, were conducted with supporting legal conditions and corresponding operating conditions to propose appropriate designs for the SH for self-employed people in HCMC. First, observing and analyzing urban spaces helped identify the unused urban areas that solve the investment issue. Then, after studying the development of social housing in different contexts via the questionnaire and in-depth interviews, self-employed households’ basic information and their business needs in using SH spaces were identified in some districts. Then, based on the legal framework and practical projects, optimal space designs were formed.

Architecturally safe and healthy classrooms: eco-medical concept to achieve sustainability in light of COVID-19 global pandemic

ABSTRACT In March 2020, the World Health Organization (WHO) officially announced the COVID-19 outbreak as a global pandemic. With this announcement, the world is facing a major challenge. This research commenced as an initiative to conceptualize internal architectural space design to cope with epidemics, to ensure an architectural anti-virus environment. In fact, architectural spaces with high occupancy loads are considered a fertile environment for the spread of this virus. This study discusses classrooms as one of those patterns with high densities for a significant reason; previous studies have revealed that pupils between the ages of 5 to 18 years may be carriers of the infection, but without symptoms; therefore, this category is considered a major source of transmitting the infection to their families and society. Accordingly, this research will address the most important standards, and design norms for these spaces using an electronic questionnaire and computational fluid dynamics (CFD) model, to achieve two objectives: first, a healthy classroom to reduce the COVID-19 infection among pupils to promote both wellbeing and performance of users, and second, a sustainable classroom from architecture, HVAC, and a medical point of view. In addition, four alternative classroom designs were proposed based on the interpretive results from the questionnaire and simulation.

Analysis of the Impact of In-Between Meaning on Promoting the Sense of Place in the Qajar Mosque-Schools (Case Study: Salehiye Mosque- School in Qazvin city)

Background and Objective: In Islamic architecture of Iran, public buildings such as mosques-schools were places for people to interact with different tastes and thoughts, with their diverse functions and spaces and with in-between spaces that played a decisive role in organizing the diverse functions of these places and linking them more closely. They formed a whole, and made it possible for one to stay in that place. The purpose of this study is to investigate the effect of the in-between spaces on promoting a sense of place in the Qajar mosque-schools and to explore how in-between spaces can be effective in promoting a sense of place?Methods: Correlation analysis in Spss software was used to analyze the variables, and with library, field and survey studies to investigate the effective factors in promoting sense of place and in-between spaces.Findings: Results of this study indicate that the coefficient of significance between the variables of sense of place and in-between spaces is less than 0.05 and there is a significant relationship between these two variables.Conclusion: The use of in-between spaces can be effective in creating sense of place in architectural space. Because today the concept of in-between spaces is very important in the design of architectural spaces, so that it can be used to reinforce the link between human and place in new public spaces, ultimately leading to better exploitation of the environment and the continued presence of people.

Accelerated design of architectured ceramics with tunable thermal resistance via a hybrid machine learning and finite element approach

Topologically interlocked architectures can transform brittle ceramics into tougher materials, while making the material design procedure a cumbersome task since modeling the whole architectural design space is not efficient and, to a degree, is not viable. We propose an approach to design architectured ceramics using machine learning (ML), trained by finite element analysis data and together with a self-learning algorithm, to discover high-performance architectured ceramics in thermomechanical environments. First, topologically interlocked panels are parametrically generated. Then, a limited number of designed architectured ceramics subjected to a thermal load is studied. Finally, the multilinear perceptron is employed to train the ML model in order to predict the thermomechanical performance of architectured panels with varied interlocking angles and number of blocks. The developed feed-forward artificial neural network framework can boost the architectured ceramic design efficiency and open up new avenues for controllability of the functionality for various high-temperature applications. This study demonstrates that the architectured ceramic panels with the ML-assisted engineered patterns show improvement up to 30% in frictional energy dissipation and 7% in the sliding distance of the tiles and 80% reduction in the strain energy, leading to a higher safety factor and the structural failure delay compared to the plain ceramics.

Effect of Structural Design Efficiency in Achieving Stability and Efficiency of the Architectural Design

The research dealt with the historical development of construction systems, their compatibility with architectural spaces, and the functional and expressive aspects of the various construction systems. Then, it discussed the creativity in the construction systems, which composed of structural elements, and elements interrelated and interactive with each other, in order to complete the structural function specified for design, with the provision of stability elements. The research was conducted based on the analytical method by adopting a field study and a research tool, which was the questionnaire form used in collecting data from the study sample to discover the effect of the contemporary construction systems and materials on the architectural spaces, after verifying its validity and performing the statistical analysis. The research sample was selected according to the study variables (gender, education, and years of experience). It was found that the diversity in construction materials and systems can lead to flexibility in the architectural space. Also, a comfortable space, problem-free, can be a result of the compatibility between architectural design and structural design. Furthermore, there is a functional impact of modern construction materials on space by adopting the surface texture of the construction materials to address the architectural space design.

SDCRKL-GP: Scalable deep convolutional random kernel learning in gaussian process for image recognition

Deep convolutional neural networks have shown great potential in image recognition tasks. However, the fact that the mechanism of deep learning is difficult to explain hinders its development. It involves a large amount of parameter learning, which results in high computational complexity. Moreover, deep convolutional neural networks are often limited by overfitting in regimes in which the number of training samples is limited. Conversely, kernel learning methods have a clear mathematical theory, fewer parameters, and can contend with small sample sizes; however, they are not able to handle high-dimensional data, e.g., images. It is important to achieve a performance and complexity trade-off in complicated tasks. In this paper, we propose a novel scalable deep convolutional random kernel learning in Gaussian process architecture called SDCRKL-GP, which is characterized by excellent performance and low complexity. First, we successfully incorporated the deep convolutional architecture into kernel learning by implementing the random Fourier feature transform for Gaussian processes, which can effectively capture hierarchical and local image-level features. This approach enabled the kernel method to effectively handle image processing problems. Second, we optimized the parameters of deep convolutional filters and Gaussian kernels by stochastic variational inference. Then, we derived the lower variational bound of the marginal likelihood. Finally, we explored the model architecture design space selection method to determine the appropriate network architecture for different datasets. The design space consists of the number of layers, the channels per layer, and so on. Different design space selections improved the scalability of the SDCRKL-GP architecture. We evaluated SDCRKL-GP on the MNIST, FMNIST, CIFAR10, and CALTECH4 benchmark datasets. Taking MNIST as an example, the error rate of classification is 0.60%, and the number of parameters, number of computations and memory access cost of the architecture are 19.088k, 0.984M, and 1.057M, respectively. The experimental results verified that the proposed SDCRKL-GP method outperforms several state-of-the-art algorithms in both accuracy and speed in image recognition tasks. The code is available at https://github.com/w-tingting/deep-rff-pytorch.

Dynamically Grown Generative Adversarial Networks

Recent work introduced progressive network growing as a promising way to ease the training for large GANs, but the model design and architecture-growing strategy still remain under-explored and needs manual design for different image data. In this paper, we propose a method to dynamically grow a GAN during training, optimizing the network architecture and its parameters together with automation. The method embeds architecture search techniques as an interleaving step with gradient-based training to periodically seek the optimal architecture-growing strategy for the generator and discriminator. It enjoys the benefits of both eased training because of progressive growing and improved performance because of broader architecture design space. Experimental results demonstrate new state-of-the-art of image generation. Observations in the search procedure also provide constructive insights into the GAN model design such as generator-discriminator balance and convolutional layer choices.

Analysis of Proposed Adaptation of Fostered Environment and Evaluation of Built Environment in Bali in Facing Covid-19

At the time this paper started being written, the spread of the COVID-19 pandemic had been taking place for more than a year. To date, it has caused fear in society because the number of deaths it causes is nearly 3 million people, the effectiveness of the vaccines being developed is uncertain and there is no convincing drug available. As a result, the virus having hit almost all countries and regions around the world has affected the way humans behave and will have an impact on the architectural design and urban spaces in which humans live. This paper explores the possible criteria for architectural design and urban design in response to a pandemic, particularly in Bali. Plural society serves as the characteristic of the island whose inhabitants have traditional activities mixed with modern trends. The mixed society has a rich tradition of communal living with social activities involving large masses of people. To produce post pandemic design criteria, this study critically reviewed recent publications including peer-reviewed articles, journals, and expert’s blog posts. In the next stage, the results of the review were used to evaluate activity patterns and fostered environment at several locations in Bali. The locations observed included traditional houses in Ubud, urban houses in Denpasar, offices, and traditional markets. Secondary data were organised into several themes. Meanwhile, field observations were focused on the possibility of implementing Health protocols in different places as well as community compliance in complying with them. This study found that during the period of limiting community activities due to the pandemic, health, safety and environmental hygiene had received the attention of all groups. The study also revealed that the built environment tried to make many adjustments. Nevertheless, the effectiveness of these adjustments is questionable. For this reason, further research that will produce new built spaces needs to be carried out. It will result in the need for more detailed design criteria as a mechanism to prevent the spread of the virus.

Research on the Space Design and Planning of Natural Landscape Architecture Based on Virtual Reality Technology

VR tech can build a realistic space scene, ameliorate the sense of human-computer interaction and experience, realize the effective and vivid simulation of the architectural space scene, so as to stimulate and release the imagination and creativity of designers, so it has important research value. Based on this, this paper first analyses the architecture of VR system, then studies the utilization of VR tech in natural landscape architectural space design and planning, and finally gives the process of architectural planning and design based on VR.

A multi-objective LSM/NoC architecture co-design framework

Liquid state machine(LSM) is an attractive spiking neural network (SNN) for Network-on-Chip(NoC)-based neuromorphic platforms due to their biological characteristics and hardware efficiency. But the randomly connected topology of the liquid in LSM and lots of communication spike bring different dataflow and communication congestion on the NoC-based platform. Aiming to design an accurate and communication optimized LSM architecture, we have to explore the LSM and NoC architecture design space. Enormous design space and the gap between LSM/NoC design space bring challenges to find out the optimal pair of LSM/NoC architecture design. To face the above challenge, we propose a multi-objective LSM/NoC architecture co-design framework, which fast and efficiently explores the design space of LSM/NoC to generate an optimal LSM architecture with low latency on NoC-based platform. Evaluation results show that our framework can generate LSM architecture suitable for execution on NoC-based platform with reduced runtime and negligible reduced accuracy. Compared with state-of-the-art LSM designs with the fixed NoC structure, we achieve 2.5x ∼ 3.0x latency reduction or average 3.1x energy reduction. For fair comparison, compared with state-of-the-art LSM designs with our NOC architecture search process, our framework can achieve 1.25x ∼ 1.41x lower latency and 1.16x ∼ 1.87x lower energy together with only average 0.65% accuracy loss.

Evaluating Architectural, Redundancy, and Implementation Strategies for Radiation Hardening of FinFET Integrated Circuits

In this article, authors explore radiation hardening techniques through the design of a test chip implemented in 16-nm FinFET technology, along with architectural and redundancy design space exploration of its modules. Nine variants of matrix multiplication were taped out and irradiated with neutrons. The results obtained from the neutron campaign revealed that the radiation-hardened variants present superior resiliency when either local or global triple modular redundancy (TMR) schemes are employed. Furthermore, simulation-based fault injection was utilized to validate the measurements and to explore the effects of different implementation strategies on failure rates. We further show that the interplay between these different implementation strategies is not trivial to capture and that synthesis optimizations can effectively break assumptions about the effectiveness of redundancy schemes.

Landscape Design System of Industrial Heritage Based on Big Data and web Technology

With the rapid transformation of industrial structure, most of the abandoned old industrial buildings left over from history have been abandoned, which is difficult to get people’s attention. However, industrial heritage has important aesthetic, historical and architectural heritage value. In the transformation of parks around the world, they often attach importance to their industrial layout and architectural landscape space design. The research on public commercial supporting space starts late and the starting point is relatively low. As a result, the construction of commercial facilities in China is not mature and perfect, lack of relevant basic commercial supporting materials. Therefore, how to improve the reuse rate of industrial waste, realize the continuation of industrial memory, improve the construction of public facilities, and meet all kinds of conditions required by tourists has become an important research topic. The purpose of this project is to make a reasonable allocation of landscape space for a certain place in our city. The results show that 12 people over 45 years of age are very dissatisfied with the design of industrial heritage.

The Cognitive-Emotional Design and Study of Architectural Space: A Scoping Review of Neuroarchitecture and Its Precursor Approaches.

Humans respond cognitively and emotionally to the built environment. The modern possibility of recording the neural activity of subjects during exposure to environmental situations, using neuroscientific techniques and virtual reality, provides a promising framework for future design and studies of the built environment. The discipline derived is termed “neuroarchitecture”. Given neuroarchitecture’s transdisciplinary nature, it progresses needs to be reviewed in a contextualised way, together with its precursor approaches. The present article presents a scoping review, which maps out the broad areas on which the new discipline is based. The limitations, controversies, benefits, impact on the professional sectors involved, and potential of neuroarchitecture and its precursors’ approaches are critically addressed.

A neural architecture search based framework for liquid state machine design

Liquid state machines (LSMs), also known as the recurrent version of spiking neural networks, have garnered significant research interest owing to their high computational power, biological plausibility, simple structure, and low training complexity. By exploring the design space in network architectures and parameters, recent works have demonstrated the great potential for improving the accuracy of LSM models with low complexity. However, these works are based on manually defined network architectures or predefined parameters, which may ignore the potential optimization of the architectures and parameters of LSMs. In this study, we propose a neural architecture search-based framework to explore the architecture and parameter design space for the automatic dataset-oriented LSM models. To manage the exponentially increasing design space, we adopt a three-step search for LSMs, including dynamic multiple-liquid architecture search in multiple layers, variations in the number of neurons in each liquid, and parameter search such as percentage connectivity and excitatory neuron ratio within each liquid. In addition, we propose the use of a simulated annealing algorithm to implement three-step heuristic search. Two datasets, including the image dataset of NMNIST and speech dataset of FSDD, were used to test the effectiveness of the proposed framework. Simulation results demonstrated that our framework can produce the dataset-oriented optimal LSM models with high accuracy and low complexity. The best classification accuracy on the two datasets with only 1000 spiking neurons was observed to be 92.5% and 84%. Meanwhile, the network connections of discovered optimal multiple-liquid LSM models for the two datasets, on average, were reduced by 56.3% and 60.2% separately compared with a single LSM. Furthermore, the total number of neurons in the optimal multiple-liquid LSM models on the two datasets was reduced by 20% with an accuracy loss of only 0.5%.