Innovative Architecture Research Articles

Wideband image-reject RF channelization based on soliton microcombs (invited paper)

Wideband radio frequency (RF) channelization is essential for the reception and detection of cross-band RF signals in various applications, including communications, radar, and spectrum sensing. However, digital channelizers are inefficient at performing RF channelization over a working bandwidth above 10 GHz. Meanwhile, current photonic RF channelizers face challenges in simultaneously considering a wideband, multi-channel, and a high crosstalk suppression ratio. In this work, we proposed and demonstrated a wideband image-reject RF channelization scheme based on integrated dual-soliton microcombs. The dual-soliton microcombs are used for RF spectral copies and heterodyne detection, respectively. Supported by image-reject mixers, the RF channelization is verified with an 8–37 GHz working bandwidth, a 1.2 GHz channel bandwidth, and 25 channels. The image suppression ratio is higher than 34 dB for single-tone signals and 20 dB for wideband signals. Our approach provides an innovative architecture of integrated photonic RF channelizers with high performance, which can benefit a wide range of RF applications by miniaturizing the systems.

Unlocking COVID-19 patterns: Exploring deep learning models for precise recognition and classification of CT images

Coronavirus is a pestilence sickness that truly affects old individuals and patients with persistent illnesses and causes life threatening diseases. Preventing the entire world from this epidemic, quick and accurate detection of COVID-19 plays a crucial role. To contain the advancement of Covid, it is important to build up a dependent and fast technique to recognize the individuals who are influenced and segregate them until full recovery is made. In this study, we introduce a groundbreaking deep CNN model, leveraging the latest advancements in the field, to accomplish precise categorization of COVID-19 CT images. We employ the publicly available HUST-19 dataset, encompassing an impressive collection of 13,980 CT scan images. By harnessing the power of this extensive dataset, our model aims to improve the precision and reliability of COVID-19 classification. In our research, we propose three innovative architectures for deep convolutional neural networks (CNNs), known as AlexNet, Inception V3 and VGG19 models, to get the accurate diagnosis of COVID-19 patients based on images obtained from a CT scan. In order to assess the effectiveness of CNN-based models in a quantitative manner, we employ several key metrics, including Accuracy, Precision, Recall and F-score. The findings of our study indicate that all these three architectures gave remarkable results but InceptionV3 came on the top with testing accuracy of 99.95% with precision, recall and f1-score of 1, 1, and 1 respectively. The results underscore the potential of our suggested method in accurately diagnosing COVID-19 cases utilizing CT scan images.

Use of Ensemble Learning to Improve Performance of Known Convolutional Neural Networks for Mammography Classification

Convolutional neural networks and deep learning models represent the gold standard in medical image classification. Their innovative architectures have led to notable breakthroughs in image classification and feature extraction performance. However, these advancements often remain underutilized in the medical imaging field due to the scarcity of sufficient labeled data which are needed to leverage these new features fully. While many methodologies exhibit stellar performance on benchmark data sets like DDSM or Minimias, their efficacy drastically decreases when applied to real-world data sets. This study aims to develop a tool to streamline mammogram classification that maintains high reliability across different data sources. We use images from the DDSM data set and a proprietary data set, YERAL, which comprises 943 mammograms from Mexican patients. We evaluate the performance of ensemble learning algorithms combined with prevalent deep learning models such as Alexnet, VGG-16, and Inception. The computational results demonstrate the effectiveness of the proposed methodology, with models achieving 82% accuracy without overtaxing our hardware capabilities, and they also highlight the efficiency of ensemble algorithms in enhancing accuracy across all test cases.

INCHAIN: a cyber insurance architecture with smart contracts and self-sovereign identity on top of blockchain

AbstractDespite the rapid growth of the cyber insurance market in recent years, insurance companies in this area face several challenges, such as a lack of data, a shortage of automated tasks, increased fraudulent claims from legal policyholders, attackers masquerading as legal policyholders, and insurance companies becoming targets of cybersecurity attacks due to the abundance of data they store. On top of that, there is a lack of Know Your Customer procedures. To address these challenges, in this article, we present , an innovative architecture that utilizes Blockchain technology to provide data transparency and traceability. The backbone of the architecture is complemented by Smart Contracts, which automate cyber insurance processes, and Self-Sovereign Identity for robust identification. The effectiveness of ’s architecture is compared with the literature against the challenges the cyber insurance industry faces. In a nutshell, our approach presents a significant advancement in the field of cyber insurance, as it effectively combats the issue of fraudulent claims and ensures proper customer identification and authentication. Overall, this research demonstrates a novel and effective solution to the complex problem of managing cyber insurance, providing a solid foundation for future developments in the field.

VBQ-Net: A Novel Vectorization-Based Boost Quantized Network Model for Maximizing the Security Level of IoT System to Prevent Intrusions

Data sharing with additional devices across wireless networks is made simple and advantageous by the Internet of Things (IoT), an emerging technology. However, IoT systems are more susceptible to cyberattacks because of their continued growth and technological advances, which could lead to powerful assaults. An intrusion detection system is one of the key defense mechanisms for information and communications technology. The primary shortcomings that plague current IoT security frameworks are their inability to detect intrusions properly, their substantial latency, and their prolonged processing time and delay. Therefore, this work develops a clever and innovative security architecture called Vectorization-Based Boost Quantized Network (VBQ-Net) for protecting IoT networks. Here, a Vector Space Bag of Words (VSBW) methodology is used to reduce the dimensionality of features and identify a key characteristic from the featured data. In addition, a brand-new classification technique, called Boosted Variance Quantization Neural Networks (BVQNNs), is used to classify the different types of intrusions using a weighted feature matrix. A Multi-Hunting Reptile Search Optimization (MH-RSO) algorithm is employed during categorization to calculate the probability value for selecting the right choices while anticipating intrusions. In this study, the most well-known and current datasets, such as IoTID-20, IoT-23, and CIDDS-001, are used to validate and evaluate the effectiveness of the proposed methodology. By evaluating the proposed approach on standard IoT datasets, the study seeks to address the limitations of current IoT security frameworks and provide a more effective defense mechanism against cyberattacks on IoT systems.

Efficient Harris Hawk Optimization (HHO)-Based Framework for Accurate Skin Cancer Prediction

The prediction of skin cancer poses a number of challenges due to the differences in visual characteristics between melanoma, basal cell carcinomas, and squamous cell carcinomas. These visual differences pose difficulties for models in discerning subtle features and patterns accurately. However, a remarkable breakthrough in image analysis using convolutional neural networks (CNNs) has emerged, specifically in the identification of skin cancer from images. Unfortunately, manually designing such neural architectures is prone to errors and consumes substantial time. It has become increasingly popular to design and fine-tune neural networks by using metaheuristic algorithms that are based on natural phenomena. A nature-inspired algorithm is a powerful alternative to traditional algorithms for solving problems, particularly in complex optimization tasks. One such algorithm, the Harris hawk optimization (HHO), has demonstrated promise in automatically identifying the most appropriate solution across a wide range of possibilities, making it suitable for solving complex optimization problems. The purpose of this study is to introduce a novel automated architecture called “HHOForSkin” that combines the power of convolutional neural networks with meta-heuristic optimization techniques. The HHOForSkin framework uses an innovative custom CNN architecture with 26 layers for the analysis of medical images. In addition, a Harris hawk optimization algorithm (HHO) is used to fine-tune the developed model for multiple skin cancer classification problems. The developed model achieves an average accuracy of 99.1% and 98.93% F1 score using a publicly available skin cancer dataset. These results position the developed optimization-based skin cancer detection strategy at the forefront, offering the highest accuracy for seven-class classification problems compared to related works.

A Microservices-Based Approach to Designing an Intelligent Railway Control System Architecture

The symmetry between customer expectations and operator goals, on one hand, and the digital transition of the railways, on the other hand, is one of the main factors affecting green transport sustainability. The European Train Control System (ETCS) was created to improve interoperability between different railway signaling systems and increase safety and security. While there are a lot of ETCS Level 2 deployments all over the world, the specifications of ETCS Level 3 are under development. ETCS Level 3 is expected to have a significant impact on automatic train operation, protection, and supervision. In this paper, we present an innovative control system architecture that allows the incorporation of artificial intelligence (AI)/machine learning (ML) applications. The architecture features control function virtualization and programmability. The concept of an intelligent railway controller (IRC) is introduced as being a piece of cloud software responsible for the control and optimization of railway operations. A microservices-based approach to designing the IRC’s functionality is presented. The approach was formally verified, and some of its performance metrics were identified.

A Novel Framework of Genetic Algorithm and Spectre to Optimize Delay and Power Consumption in Designing Dynamic Comparators

In integrated circuit (IC) design, analog circuits contribute significantly as the interface between real and digital world signals. Although they make up a relatively small portion of the overall circuit, their design process is often most time-consuming, mostly from the phase of manual iteration of circuit parameters to meet design specifications. Therefore, the design automation of analog circuits with the help of efficient optimization techniques arises as a promising candidate to address the issue. Among optimization algorithms, while the genetic algorithm (GA) has been shown to be effective in finding near-optimal solutions, it has not been extensively applied to the field of analog circuit design. Hence, this paper proposes a method to utilize GA in the optimization of a widely used circuit topology, namely the comparator. The comparator is considered the fundamental block in the design of most analog-to-digital converters (ADCs). For high-speed ADCs, dynamic comparators are usually chosen for the purpose of high speed and power efficiency. In summary, this paper introduces an innovative GA-Spectre architecture to optimize the dynamic comparator with respect to delay and power consumption. The post-optimized results are optimistic with a 72.61 ps delay and 3.11 µW power dissipation.

BiFTransNet: A unified and simultaneous segmentation network for gastrointestinal images of CT & MRI

Gastrointestinal (GI) cancer is a malignancy affecting the digestive organs. During radiation therapy, the radiation oncologist must precisely aim the X-ray beam at the tumor while avoiding unaffected areas of the stomach and intestines. Consequently, accurate, automated GI image segmentation is urgently needed in clinical practice. While the fully convolutional network (FCN) and U-Net framework have shown impressive results in medical image segmentation, their ability to model long-range dependencies is constrained by the convolutional kernel’s restricted receptive field. The transformer has a robust capacity for global modeling owing to its inherent global self-attention mechanism. The TransUnet model leverages the strengths of both the convolutional neural network (CNN) and transformer models through a hybrid CNN-transformer encoder. However, the concatenation of high- and low-level features in the decoder is ineffective in fusing global and local information. To overcome this limitation, we propose an innovative transformer-based medical image segmentation architecture called BiFTransNet, which introduces a BiFusion module into the decoder stage, enabling effective global and local feature fusion by enabling feature integration from various modules. Further, a multilevel loss (ML) strategy is introduced to oversee the learning process of each decoder layer and optimize the use of globally and locally fused contextual features at different scales. Our method achieved a Dice score of 89.51% and an intersection-over-union (IoU) score of 86.54% on the UW-Madison Gastrointestinal Segmentation dataset. Moreover, our method attained a Dice score of 78.77% and a Hausdorff distance (HD) of 27.94% on the Synapse Multi-organ Segmentation dataset. Compared with the state-of-the-art methods, our proposed method achieves superior segmentation performance in gastrointestinal segmentation tasks. More significantly, our method can be easily extended to medical segmentation in different modalities such as CT and MRI. Our method achieves clinical multimodal medical segmentation and provides decision supports for clinical radiotherapy plans.

AttentionCode: Ultra-Reliable Feedback Codes for Short-Packet Communications

Ultra-reliable short-packet communication is a major challenge in future wireless networks with critical applications. To achieve ultra-reliable communications beyond 99.999%, this paper envisions a new interaction-based communication paradigm that exploits feedback from the receiver. We present AttentionCode, a new class of feedback codes leveraging deep learning (DL) technologies. The underpinnings of AttentionCode are three architectural innovations: AttentionNet, input restructuring, and adaptation to fading channels, accompanied by several training methods, including large-batch training, distributed learning, look-ahead optimizer, training-test signal-to-noise ratio (SNR) mismatch, and curriculum learning. The training methods can potentially be generalized to other wireless communication applications with machine learning. Numerical experiments verify that AttentionCode establishes a new state of the art among all DL-based feedback codes in both additive white Gaussian noise (AWGN) channels and fading channels. In AWGN channels with noiseless feedback, for example, AttentionCode achieves a block error rate (BLER) of 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-7</sup> when the forward channel SNR is 0 dB for a block size of 50 bits, demonstrating the potential of AttentionCode to provide ultra-reliable short-packet communications.

Receiving and Transmitting Antenna Pairs Based Rasorber With High Angular Stability

Frequency selective rasorbers (FSRs) with out-of-band absorption and in-band transmission performance have been widely reported for stealth radome applications over the past few years. However, developing an FSR with high angular stability in the transmission band remains challenging. Therefore, to improve the transmission angular stability, we propose an innovative architecture in this work for constructing the FSR with independently explorable absorption and transmission paths. First, a band-stop frequency selective surface (FSS) and an absorber are employed in the absorption path to obtain out-of-band absorption. Second, the receiving (RX) and transmitting (TX) antenna pairs connected by the corresponding enclosed transmission lines (TLs) are utilized to comprise the transmission path. Note that a wide-beam RX antenna is employed to capture incident electromagnetic (EM) waves at large angles and achieve in-band transmission with high angular stability. Besides, another second-order FSR with a sharp roll-off transmission filtering response is proposed by using the second-order band-stop FSS as the ground (GND) of the RX antenna. Furthermore, two prototypes of the FSRs (one narrowband and the other second-order) are fabricated and measured as demonstrations of the proposed architecture. Experimental results agree with simulation ones, indicating the two devices exhibit the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$-$</tex-math> </inline-formula> 3-dB transmission bands with excellent angular stability of 60 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$^\circ$</tex-math> </inline-formula> , and both transmission windows are located on a wide <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$-$</tex-math> </inline-formula> 10-dB low reflection band.

Development and Validation of a Smart Architecture for Thyristor Valves

A high voltage thyristor converter is realized by many valve sections, whose volume is approximately occupied for only the 10% by thyristors and for the 10% by the relevant gate drivers. The remaining 80% is taken by passive auxiliary circuits, needed to protect thyristors during turn-on and turn-off commutations. This work represents a preliminary validation of an innovative architecture that aims to reduce the auxiliary circuit cost, volume, and weight of the overall valve, through the investigation of active, instead of passive solutions. The work starts from investigation of the phenomena behind the valve transient behaviors and proceeds with simulations and experimental tests, using a reduced scale circuit prototype. The match between the obtained results validates the investigated active configurations, confirming that the proposed solution can be used for improving thyristor valves.

Carbon Nanorings and Nanobelts: Material Syntheses, Molecular Architectures, and Applications

AbstractCarbon nanorings (CNRs) and carbon nanobelts (CNBs) of various sizes and innovative molecular architectures have ushered improvement in research in recent years. This serves as a prerequisite for further advances and applications, such as organic semiconductors and bottom‐up synthesis of single‐walled carbon nanotubes. Nevertheless, challenges such as high strain energy, excessive reactivity, end‐product stability, low‐scale yields, and side reactions inhibit material synthesis and applications. In this review, CNRs, CNBs, and other heteroatom‐doped molecular belts are discussed based on their molecular structural characteristics, with a special focus on their developments in the past two years. Furthermore, current applications, research obstacles, and future developments related to CNRs and CNBs are discussed. For the first time, studies and advancements in organic field‐effect transistors (OFETs) have been summarized in detail.

Cloud‐Enabled Handheld NIR Spectroscopy: A Transformative Approach for Real‐Time Forensic Analysis of Cannabis Specimens

AbstractIn the past few years, there has been significant interest within the forensic community regarding the deployment of portable solutions that provide real‐time results. This article introduces an innovative technology or technology architecture that enables the integration of a handheld device, specifically, Viavi MicroNIR, with a cloud‐based system. This cloud system encompasses a server responsible for data processing and a mobile application acting as a user interface.To demonstrate the transformative impact of this technology on field operators, the analysis of cannabis specimens has been utilized. System's capacity to distinguish between CBD‐type and THC‐type cannabis has been particularly highlighted, along with the remarkable congruence observed between the near‐infrared (NIR) spectra and the reference analytical method involving ultra‐high‐performance liquid chromatography (UHPLC)The article will present the advantages of this application primarily focusing on its potential to alleviate the burden on laboratories by expediting routine illicit drug analysis. Viavi MicroNIR technology provides laboratory personnel with additional time to handle more complex cases, thereby enhancing overall efficiency.

Next-generation Distributed Storage Technologies: Architectural Innovations and Performance Analysis

Next-generation distributed storage technologies have emerged as critical components in modern data management systems, offering scalability, performance, and security for handling vast amounts of data. This research paper explores various aspects of these technologies, including architectural innovations, performance analysis, real-world deployments, security considerations, and future trends. Through an examination of case studies and industry reports, the paper highlights the practical applications and benefits of distributed storage solutions in diverse sectors such as e-commerce, healthcare, and information technology. Additionally, the paper discusses key factors such as data consistency, integrity, energy efficiency, and environmental impact, shedding light on the challenges and opportunities associated with distributed storage systems. By analysing numerical data and industry insights, the paper provides valuable insights into the current state and future trajectory of distributed storage technologies. Overall, this comprehensive study aims to contribute to the understanding and advancement of next-generation distributed storage systems in the digital age.

Robust Reinforcement Learning-Based Multiple Inputs and Multiple Outputs Controller for Wind Turbines

The control of variable-speed wind turbines that generate electricity from the kinetic energy of the wind involves subsystems that need to be controlled simultaneously, namely, the blade pitch angle controllers and the generator torque controllers. The presented study solves the control problem with multiple inputs and multiple outputs (MIMO), using the method of reinforcement learning–based Trust Region Policy Optimization, through which the control parameters of both subsystems are simultaneously optimized. In this case, the robust control problem is transformed into a constrained optimal control problem with an appropriate choice of value functions for the nominal system. The study aims to synthesize a robust controller, with the aim of maximizing the generated energy (power) and minimizing unwanted forces (thrust). The innovative control architecture uses an extended input space, which allows fine-tuning of parameters for each operating state. Test calculations carried out in simulation experiments using models of the 5 MW NREL wind turbine and the 4 MW Enercon E-126 EP3 wind turbine are presented to illustrate the performance and practicality of the proposed approach.

Assessment of emerging pretraining strategies in interpretable multimodal deep learning for cancer prognostication

BackgroundDeep learning models can infer cancer patient prognosis from molecular and anatomic pathology information. Recent studies that leveraged information from complementary multimodal data improved prognostication, further illustrating the potential utility of such methods. However, current approaches: 1) do not comprehensively leverage biological and histomorphological relationships and 2) make use of emerging strategies to “pretrain” models (i.e., train models on a slightly orthogonal dataset/modeling objective) which may aid prognostication by reducing the amount of information required for achieving optimal performance. In addition, model interpretation is crucial for facilitating the clinical adoption of deep learning methods by fostering practitioner understanding and trust in the technology.MethodsHere, we develop an interpretable multimodal modeling framework that combines DNA methylation, gene expression, and histopathology (i.e., tissue slides) data, and we compare performance of crossmodal pretraining, contrastive learning, and transfer learning versus the standard procedure.ResultsOur models outperform the existing state-of-the-art method (average 11.54% C-index increase), and baseline clinically driven models (average 11.7% C-index increase). Model interpretations elucidate consideration of biologically meaningful factors in making prognosis predictions.DiscussionOur results demonstrate that the selection of pretraining strategies is crucial for obtaining highly accurate prognostication models, even more so than devising an innovative model architecture, and further emphasize the all-important role of the tumor microenvironment on disease progression.

A Framework for Representing, Building and Reusing Novel State-of-the-Art Three-Dimensional Object Detection Models in Point Clouds Targeting Self-Driving Applications.

The rapid development of deep learning has brought novel methodologies for 3D object detection using LiDAR sensing technology. These improvements in precision and inference speed performances lead to notable high performance and real-time inference, which is especially important for self-driving purposes. However, the developments carried by these approaches overwhelm the research process in this area since new methods, technologies and software versions lead to different project necessities, specifications and requirements. Moreover, the improvements brought by the new methods may be due to improvements in newer versions of deep learning frameworks and not just the novelty and innovation of the model architecture. Thus, it has become crucial to create a framework with the same software versions, specifications and requirements that accommodate all these methodologies and allow for the easy introduction of new methods and models. A framework is proposed that abstracts the implementation, reusing and building of novel methods and models. The main idea is to facilitate the representation of state-of-the-art (SoA) approaches and simultaneously encourage the implementation of new approaches by reusing, improving and innovating modules in the proposed framework, which has the same software specifications to allow for a fair comparison. This makes it possible to determine if the key innovation approach outperforms the current SoA by comparing models in a framework with the same software specifications and requirements.

Digital Fabrication in Architecture

Digital fabrication has emerged as a transformative force in the field of architecture, revolutionizing design, and construction processes. The paper begins by providing an overview of digital fabrication and its evolution within the architectural domain. It discusses the integration of advanced technologies such as computer-aided design (CAD), additive manufacturing (3D printing) into architectural processes. The advantages of digital fabrication, including increased design flexibility, enhanced precision, reduced material waste, and accelerated construction timelines, are highlighted. Furthermore, the paper investigates various case studies showcasing the successful implementation of digital fabrication in architecture. Through these examples, the potential of digital fabrication techniques to enable complex geometries and facilitate sustainable design practices is underscored. To conclude, this paper emphasizes the immense potential of digital fabrication as a tool for architectural innovation. It advocates for further research and development in this field to overcome the challenges and maximize the benefits. By embracing digital fabrication, architects can push the boundaries of design, create structures of unparalleled intricacy, and achieve sustainable and efficient built environments.

Preparation of CuMn2O4/Ti3C2 MXene composite electrodes for supercapacitors with high energy density and study on their charge transfer kinetics

In this study, we present a novel electrode material that combines Ti3C2 MXene and high-capacity CuMn2O4 to increase the energy density of supercapacitors, which are a popular choice for energy storage due to their high-performance potential. The electrode material was synthesized using the hydrothermal method with varying deposition times (3 h, 6 h and 9 h), and the resulting composite materials were characterized using advanced analytical techniques. The CuMn2O4/MXene composite electrode synthesized at 3h exhibited exceptional performance, with a specific capacitance of 628 mF/cm2 at 4 mA/cm2, due to the enhanced electrical conductivity and charge storage properties of CuMn2O4 and MXene sheets. We also uncovered an intricate charge transfer mechanism and storage kinetics of CuMn2O4/MXene composite on a nickel foam electrode, revealing a diffusion-controlled energy storage mechanism with fast mass transportation. To demonstrate practicality, we constructed an asymmetric coin cell supercapacitor device using CuMn2O4/MXene composite synthesized at 3h and activated carbon as the positive and negative electrodes, respectively. The device showed a specific capacitance of 496 mF/cm2 at 6 mA/cm2 with cyclic stability of 80% for up to 10,000 cycles, and a power density of 1.5 mW/cm2 at a higher energy density of 0.073 mWh/cm2. Our results demonstrate the potential to significantly advance the development of high-performance supercapacitors by combining Ti3C2 MXene and high-capacity oxides, refining the synthesis process, and exploring innovative electrode architectures.