Decomposition Architecture Research Articles

Data fusion integrated network forecasting scheme classifier (DFI-NFSC) via multi-layer perceptron decomposition architecture

The Massive Access Problem of the Internet of Things stands for the access problem of the wireless devices to the Gateway when the device population in the coverage area is excessive. We develop a hybrid model called Data Fusion Integrated Network Forecasting Scheme Classifier (DFI-NFSC) using a Multi-Layer Perceptron (MLP) Decomposition architecture specifically designed to address the Massive Access Problem. We utilize our custom error metric to display throughput and energy consumption results. These results are obtained by emulating the Joint Forecasting–Scheduling (JFS) system on a single IoT Gateway and distinguishing between ARIMA, LSTM, and MLP forecasters of the JFS system. The outcomes indicate that the DFI-NFCS method plays a notable role in improving performance and mitigating challenges arising from the dynamic fluctuations in the diversity of device types within an IoT gateway’s coverage zone.

A sequential multi-agent reinforcement learning framework for different action spaces

In many real-world decision-making tasks, multi-agent need to learn collaboration in a high-dimensional complex action space, rather than just a single discrete action space. Recently, value decomposition learning methods such as QMIX have emerged as a promising approaches for collaborative multi-agent tasks. However, most of the value decomposition algorithms can only be used in discrete action space, which would limit their practicability. To address the limitation, we propose a novel algorithm called Multi-Agent Sequential Q-Networks (MASQN), which can be applied to the multi-agent domains with continuous, multidiscrete or hybrid action spaces. The proposed algorithm is based on the structure of centralized training with decentralized execution (CTDE). The decentralized actors ensure adaptability to different action spaces by utilizing action space discretization and sequential models, and the centralized critic utilizes the value decomposition architecture to guide effective updates of the policy parameters for each agent. We also give the convergence of joint policy from the perspective of policy iteration, by combining it with the CTDE structure and the constraint of the Individual Global Max (IGM) condition. Finally, we evaluate the MASQN algorithm on two benchmark environments: MAMuJoCo and Hybrid Predator–Prey. The empirical results show that MASQN out performs the state-of-the-art performance on three different action spaces.

Ontological model of structure and parameters of components in high-tech product modernization projects

The subject of the article is models representing the structure and parameters of components in high-tech products based on the analysis of technical documentation in modernization projects. The purpose of the proposed research is to enhance the quality of modernization processes of high-tech products by forming an ontological model of the product with innovative components, considering the diversity of information support. The article addresses the following tasks: studying the main methods and technologies of system representation of complex product structure; forming a model of structural-functional decomposition of a high-tech product, creating an ontological model of the structure and parameters of a high-tech product based on technical documentation. The following methods are applied: systemic approach, methods of functional-structural decomposition, set theory, ontology construction methods, semantic models. The following results were obtained: The main methods of system representation of complex product structure were investigated, based on the following principles: decomposition of complex product architecture, layering of complex product representation, multivariate synthesis of component architecture. The main directions and advantages of using 3D technologies for solving design tasks in modernization projects of high-tech product components are considered. The decomposition of a high-tech product into component parts and partial parameters is proposed, taking into account functional, structural, and parametric characteristics. An ontological model of the structure and parameters of a high-tech product is formed based on a set of technical documentation and considering additional heterogeneous sources of information. Innovative elements are identified, the description of which may be fuzzy. Conclusions. The proposed ontological model can serve as a basis for finding similar solutions regarding the design of innovative components in the precedent database, which is properly structured. In the absence of similar design solutions, innovative component design can be carried out using 3D technologies based on supplementing fuzzy information in the semantic model.

RADF: Architecture decomposition for function as a service

AbstractAs the most successful realization of serverless, function as a service (FaaS) brings in a novel cloud computing paradigm that can save operating costs, reduce management effort, enable seamless scalability, and augment development productivity. Migration of an existing application to the serverless architecture is, however, an intricate task as a great number of decisions need to be made along the way. We propose in this paper RADF, a semi‐automatic approach that decomposes a monolith into serverless functions by analyzing the business logic inherent in the interface of the application. The proposed approach adopts a two‐stage refactoring strategy, where a coarse‐grained decomposition is performed at first, followed by a fine‐grained one. As such, the decomposition process is simplified into smaller steps and adaptable to generate a solution at either microservice or function level. We have implemented RADF in a holistic DevOps methodology and evaluated its capability for microservice identification and feasibility for code refactoring. In the evaluation experiments, RADF achieves lower coupling and relatively balanced cohesion, compared to previous decomposition approaches.

An architecture decomposition method of pneumatic catapult system based on OPM and DSM

The architecture strategy of the Unmanned Aerial Vehicle (UAV) pneumatic launch system should continue to evolve to adapt to complex and variable operating environments. Architecture representation, decomposition perspective, and cluster analysis play a vital role in the early phase of system architecture development. In order for the system to emerge anticipated and desirable intrinsic functional properties, an architecture decomposition method based on the Object-Process Methodology (OPM) and Design Structure Matrix (DSM) is put forward in this paper. The OPM is proposed to model the UAV launch process formally, and the matrix representation of the architecture of the pneumatic launch system is established. After the extension of the definition and operations of DSM, with the Idicula-Gutierrez-Thebeau Algorithm plus (IGTA + ) clustering algorithm, the transformation of the pneumatic launch system architecture from process decomposition to function decomposition is demonstrated in this paper. The analysis shows that the architecture decomposition of the pneumatic launch system meets the functional requirements of stakeholders.

Time Series Analysis Based on Informer Algorithms: A Survey

Long series time forecasting has become a popular research direction in recent years, due to the ability to predict weather changes, traffic conditions and so on. This paper provides a comprehensive discussion of long series time forecasting techniques and their applications, using the Informer algorithm model as a framework. Specifically, we examine sequential time prediction models published in the last two years, including the tightly coupled convolutional transformer (TCCT) algorithm, Autoformer algorithm, FEDformer algorithm, Pyraformer algorithm, and Triformer algorithm. Researchers have made significant improvements to the attention mechanism and Informer algorithm model architecture in these different neural network models, resulting in recent approaches such as wavelet enhancement structure, auto-correlation mechanism, and depth decomposition architecture. In addition to the above, attention algorithms and many models show potential and possibility in mechanical vibration prediction. In recent state-of-the-art studies, researchers have used the Informer algorithm model as an experimental control, and it can be seen that the algorithm model itself has research value. The informer algorithm model performs relatively well on various data sets and has become a more typical algorithm model for time series forecasting, and its model value is worthy of in-depth exploration and research. This paper discusses the structures and innovations of five representative models, including Informer, and reviews the performance of different neural network structures. The advantages and disadvantages of each model are discussed and compared, and finally, the future research direction of long series time forecasting is discussed.

Wavenumber-Division Multiplexing in Line-of-Sight Holographic MIMO Communications

Starting from first principles of wave propagation, we consider a multiple-input multiple-output (MIMO) representation of a communication system between two spatially-continuous volumes. This is the concept of holographic MIMO communications. The analysis takes into account the electromagnetic interference, generated by external sources, and the constraint on the physical radiated power. The electromagnetic MIMO model is particularized for a pair of parallel line segments in line-of-sight conditions. Inspired by orthogonal-frequency division-multiplexing, we assume that the spatially-continuous transmit currents and received fields are represented using the Fourier basis functions. In doing so, a wavenumber-division multiplexing (WDM) scheme is obtained, which is not optimal but can be efficiently implemented. The interplay among the different system parameters (e.g., transmission range, wavelength, and sizes of source and receiver) in terms of number of communication modes and level of interference among them is studied with conventional tools of linear systems theory. Due to the non-finite support (in the spatial domain) of the electromagnetic channel, WDM cannot provide non-interfering communication modes. The interference decreases as the receiver size grows, and goes to zero only asymptotically. Different digital processing architectures, operating in the wavenumber domain, are thus used to deal with the interference. The simplest implementation provides the same spectral efficiency of a singular-value decomposition architecture with water-filling when the receiver size is comparable to the transmission range. The developed framework is also used to represent a communication scheme that performs only an integration over short spatial segments. This is equivalent to a classical MIMO system with uniform linear arrays made of electrically small dipoles. Numerical comparisons show that better performance than WDM can be achieved only when a higher number of radio-frequency chains is used.

Spatio-temporal hierarchical MLP network for traffic forecasting

Traffic forecasting is an indispensable part of intelligent transportation systems. However, existing methods suffer from limited capability in capturing hierarchical temporal characteristics of the traffic time series. To be specific, they neglect the property that the time series is composed of trend-cyclical and seasonal parts. On the other hand, prior methods ignore the natural hierarchical structure of traffic road networks and thus fail to capture the macro spatial dependence of region networks. To address these issues, we propose a novel spatio-temporal hierarchical MLP network (STHMLP) for traffic forecasting. By adopting a decomposition architecture in the STHMLP, trend-cyclical and seasonal features are gradually grasped from multi-scale local compositions of traffic time series. For each scale of traffic time series, we design a fine module and a coarse module to extract spatio-temporal information from roads and regions, respectively. Specifically, the fine module utilizes spatial filters on the frequency domain features of traffic time series to efficiently capture fine-grained spatial dependencies. The coarse module adaptively coarsens road networks to region networks and captures coarse-grained spatial dependencies from region networks. Experiments on four real-world traffic datasets demonstrate the STHMLP outperforms state-of-the-art baselines on traffic forecasting.

SES-X: A MBSE Methodology Based on SES/MB and X Language

Model-based systems engineering (MBSE) is a leading paradigm for the analyses and development of complex systems. However, the development of modeling and simulation infrastructure supporting MBSE is lacking, which limits the application of MBSE. To address this problem, this paper proposes an SES-X methodology that integrates system modeling (following SES philosophy) with system simulation (supported by X language) to support the full lifecycle of MBSE modeling, including system analysis, architecture decomposition, physical modeling, and simulation. In the process, SES-X performs two levels of model pruning for model verification and simulation efficiency. This paper also conducts a case study on a car model to illustrate the effectiveness of the SES-X methodology.

Wind-Photovoltaic-Energy Storage System Collaborative Planning Strategy Considering the Morphological Evolution of the Transmission and Distribution Network

The collaborative planning of a wind-photovoltaic (PV)-energy storage system (ESS) is an effective means to reduce the carbon emission of system operation and improve the efficiency of resource collaborative utilization. In this paper, a wind-PV-ESS collaborative planning strategy considering the morphological evolution of the transmission and distribution network is proposed. Firstly, aiming at the optimal economy of transmission and distribution network and considering the constraints of safe and stable operation of the system, the planning model of the transmission network based on DC power flow and the planning model of the distribution network based on AC power flow are constructed. Further, considering the coupling interaction between the transmission and distribution networks, a collaborative planning model of transmission and distribution networks based on second-order cone relaxation (SOCR) is constructed. Secondly, in order to reduce the computational complexity of the model and ensure the global optimality of the model solution, a fast model solution method based on heterogeneous decomposition architecture is proposed. Thirdly, the multiple driving factors of the morphological evolution of transmission and distribution network are analyzed, the morphological evolution path and typical characteristics of transmission and distribution network are determined, and a wind-PV-ESS collaborative planning strategy considering the morphological evolution of a transmission and distribution network is proposed. Finally, the results show that, compared with the sprouting period, the overall economy of the development period and maturity period is improved by 3342 k$ and 5751 k$ respectively, and the effectiveness and necessity of the collaborative planning strategy proposed in this paper is verified.

Open RAN—Radio Access Network Evolution, Benefits and Market Trends

Open RAN (radio access network) movement is perceived as a game changer, having robust potential to introduce shifts in mobile radio access networks towards tailor-made solutions based on the architecture decomposition. It is widely assumed that those changes will affect the approach to network deployments and supply chains of network elements and their further integration and maintenance. First deployments of O-RAN-based networks have already delivered broadband services to end users. In parallel, many proof-of-concept feature evaluations and theoretical studies are being conducted by academia and the industry. In this review, the authors describe the RAN evolution towards open models and make an attempt to indicate potential open RAN benefits and market trends.

A novel underwater image restoration method based on decomposition network and physical imaging model

Underwater image restoration is one of the significant research in marine engineering and aquatic robotics. However, due to the propagation characteristics of light and the serious turbidity in underwater, the captured images often have chromatic aberration and scattering blur, which brings great challenges to the restoration of the raw image. In this paper, a revised underwater imaging model is proposed first, which reanalyzes the generation of background light from the atmosphere to the underwater and provides important support for underwater color correction. And then a network framework via the revised model is designed, which can decompose the captured image into different components corresponding to the revised model. The proposed network consists of a decomposition architecture with residual blocks that learns a complete separation of clear image and transmittance features. These two features are used along with the raw image to predict the background light. Finally, combining three constraints of the imaging model, the proposed framework can converge rapidly along the desired direction. By comparison with the performance of the state-of-the-art algorithms, the designed network shows excellent visibility and is capable of removing water on both synthetic and real-world images in different water types.

Large-scale and energy-efficient tensorized optical neural networks on III–V-on-silicon MOSCAP platform

This paper proposes a large-scale, energy-efficient, high-throughput, and compact tensorized optical neural network (TONN) exploiting the tensor-train decomposition architecture on an integrated III–V-on-silicon metal–oxide–semiconductor capacitor (MOSCAP) platform. The proposed TONN architecture is scalable to 1024 × 1024 synapses and beyond, which is extremely difficult for conventional integrated ONN architectures by using cascaded multi-wavelength small-radix (e.g., 8 × 8) tensor cores. Simulation experiments show that the proposed TONN uses 79× fewer Mach–Zehnder interferometers (MZIs) and 5.2× fewer cascaded stages of MZIs compared with the conventional ONN while maintaining a >95% training accuracy for Modified National Institute of Standards and Technology handwritten digit classification tasks. Furthermore, with the proven heterogeneous III–V-on-silicon MOSCAP platform, our proposed TONN can improve the footprint-energy efficiency by a factor of 1.4 × 104 compared with digital electronics artificial neural network (ANN) hardware and a factor of 2.9 × 102 compared with silicon photonic and phase-change material technologies. Thus, this paper points out the road map of implementing large-scale ONNs with a similar number of synapses and superior energy efficiency compared to electronic ANNs.

Adaptive Spatial Reasoning for Turn-Based Strategy Games

The quality of AI opponents often leaves a lot to be desired, which poses many attractive challenges for AI researchers. In this respect, Turn-based Strategy (TBS) games are of particular interest. These games are focussed on high-level decision making, rather than low-level behavioural actions. For efficiently designing a TBS AI, in this paper we propose a game AI architecture named ADAPTA (Allocation and Decomposition Architecture for Performing Tactical AI). It is based on task decomposition using asset allocation, and promotes the use of machine learning techniques. In our research we concentrated on one of the subtasks for the ADAPTA architecture, namely the Extermination module, which is responsible for combat behaviour. Our experiments show that ADAPTA can successfully learn to outperform static opponents. It is also capable of generating AIs which defeat a variety of static tactics simultaneously.

TOWARDS FUNCTIONAL ARCHITECTURAL DECOMPOSITION OF CONSTRAINED LAYER INFLATABLE SYSTEMS

AbstractPneumatically activated systems enable myriad types of highly functional inflatables employing a wide range of architectural approaches affecting their form and function, making systematic conceptual design difficult. A new architectural class of pneumatically activated systems, constrained layer inflatable systems, consists of hierarchically architected flat layers of thin airtight bladders that are internally and/or externally constrained to generate a variety of functionalities. The highly hierarchical architectural structure of constrained layer inflatable systems coincides with the hierarchy of produced functions, providing an opportunity for the development of a functional architectural decomposition, capturing the inherent relationship between architectural and functional hierarchies. The basis of the approach is conveyed through the design of an example constrained layer inflatable system. This approach empowers the systematic understanding of the interrelated architectural and functional breakdown of constrained layer inflatable systems, enabling designers to iteratively analyze, synthesize, and re-synthesize the components of the system improving existing designs and exploring new concepts.

Multiagent modeling of production logistics in the creation of high-tech products

Economic instability and stochastic behavior of the market lead to the need to develop new methods and models to reduce the life cycle of high-tech products. Therefore, the urgent task is a detailed consideration and analysis of each main stage of the life cycle of creating complex technical products. The current study develops a method for optimizing production processes. The subject of research is a production planning and management as the main stage of the life cycle. The method proposed in the publication is based on the component architecture of the created innovative high-tech product, which is obtained because of decomposition at the initial stage of design. Modern production is of distributed nature. While planning production, it is necessary to pay attention to the management and synchronization of material and other types of flows in the logistics chain of distributed production. To form a set of basic and subsidiary operations of distributed production, it is necessary to consider the logistical features, the level of decomposition of the component architecture of the created product, as well as the type of component in the production cycle. Logistics operations are conducted following the requirements for the relevant indicators, namely, the cost of the operation, quality, competitiveness, innovation, the risk of the operation. The main criterion for optimization is the duration of operations. The work uses multi-agent simulation modeling of a sequential logistics chain of production processes, as well as mathematical models and methods: system analysis, optimization using integer programming, multi-criteria optimization, and expert evaluations. The method reduces the duration of the production cycle for the manufacture of individual components of the complex architecture of high-tech products by determining the optimal set of production and ancillary logistics operations for each component, considering the requirements for key production indicators, which minimize production time of high-tech products.

Are you sure, you want a cookie? – The effects of choice architecture on users' decisions about sharing private online data

Data privacy is an important topic in the public debate. Recent policy changes aimed to strengthen users' control over their own data. We conducted a randomized field trial to show that website owners can undermine this agency by deliberately altering the choice architecture of the cookie banner and “nudge” visitors into sharing their data. Based on a sample of 1493 users, we show that cookie banner manipulations can increase consent by 17 percentage points of the sample. We propose a decomposition of the choice architecture of cookie banners into two elements: (1) the choice-making architecture and (2) the choice outcome architecture. Both can be uniquely manipulated but have different effects on privacy choices. We discuss the implications of these findings with regards to user sovereignty and the need for regulation. We conclude that the ability of website owners to manipulate the outcome of user privacy decisions is probably at odds with the ideal of the ePrivacy Directive and the GDPR.

Improved Multiview Decomposition for Single-Image High-Resolution 3D Object Reconstruction

As a representative technology of artificial intelligence, 3D reconstruction based on deep learning can be integrated into the edge computing framework to form an intelligent edge and then realize the intelligent processing of the edge. Recently, high-resolution representation of 3D objects using multiview decomposition (MVD) architecture is a fast reconstruction method for generating objects with realistic details from a single RGB image. The results of high-resolution 3D object reconstruction are related to two aspects. On the one hand, a low-resolution reconstruction network represents a good 3D object from a single RGB image. On the other hand, a high-resolution reconstruction network maximizes fine low-resolution 3D objects. To improve these two aspects and further enhance the high-resolution reconstruction capabilities of the 3D object generation network, we study and improve the low-resolution 3D generation network and the depth map superresolution network. Eventually, we get an improved multiview decomposition (IMVD) network. First, we use a 2D image encoder with multifeature fusion (MFF) to enhance the feature extraction capability of the model. Second, a 3D decoder using an effective subpixel convolutional neural network (3D ESPCN) improves the decoding speed in the decoding stage. Moreover, we design a multiresidual dense block (MRDB) to optimize the depth map superresolution network, which allows the model to capture more object details and reduce the model parameters by approximately 25% when the number of network layers is doubled. The experimental results show that the proposed IMVD is better than the original MVD in the 3D object superresolution experiment and the high-resolution 3D reconstruction experiment of a single image.

SfSNet: Learning Shape, Reflectance and Illuminance of Faces in the Wild.

We present SfSNet, an end-to-end learning framework for producing an accurate decomposition of an unconstrained human face image into shape, reflectance and illuminance. SfSNet is designed to reflect a physical lambertian rendering model. SfSNet learns from a mixture of labeled synthetic and unlabeled real-world images. This allows the network to capture low-frequency variations from synthetic and high-frequency details from real images through the photometric reconstruction loss. SfSNet consists of a new decomposition architecture with residual blocks that learns a complete separation of albedo and normal. This is used along with the original image to predict lighting. SfSNet produces significantly better quantitative and qualitative results than state-of-the-art methods for inverse rendering and independent normal and illumination estimation. We also introduce a companion network, SfSMesh, that utilizes normals estimated by SfSNet to reconstruct a 3D face mesh. We demonstrate that SfSMesh produces face meshes with greater accuracy than state-of-the-art methods on real-world images.

Blokklánc alkalmazási lehetőségek a biztonság területén és bevezetés-tervezésük

Összefoglalás. A blokklánc-technológiákat első sikeres alkalmazásuk, a kriptopénzek tették híressé és hírhedté. Valódi jelentőségük azonban az általuk létrehozott új informatikai rendszerkategóriában, az adatbázis jellegű, több résztvevő által közösen hitelesen tartott elosztott főkönyvekben rejlik. A tanulmány ismerteti ezek alapelveit, jellemző üzleti alkalmazási mintáit és a „blokkláncosítást”, mint bevezetési stratégia tervezési elvet. Új eredményként a blokkláncosítás a biztonság területén alkalmazhatóságának megteremtéséhez felállításra kerül egy ismert példákon alapuló érték-modell. Summary. Blockchain technologies were made famous – and arguably, infamous – by their first successful application: cryptocurrencies. Their true significance, however, lies in the novel IT system category they established: distributed ledgers, which are electronic systems of records maintained by multiple parties. The paper summarizes the key concepts of distributed ledger technologies, their key business application types and „blockchainification” as an innovation strategy planning methodology. As a novel contribution, the paper proposes the application of „blockchainification” in the complex context of security, and sets up an initial version of the necessary domain-specific value and application type framework. Distributed Ledger Technologies (DLT) have reached maturity where they can be applied to, and have been demonstrated to be able to, facilitate a very broad range of cross-organizational and client-organization cooperation patterns. For enterprise and industrial usage, DLT key value dimensions, supporting blockchain capbilities and value driver application types have been already collected, facilitating the structured and benefit-based planning of their introduction. One such approach is what we coined „blockchainification”. Blockchainification starts with a decomposition of the business architecture of an organization, to the point where specific cooperations can be characterized, both functionally and by the parties involved. Given such a decomposition, the viability of migrating or replacing the functionality with a DLT-based solution can be assessed, on a cooperation by cooperation basis, including the associated risks and benefits. This way, a blockchain introduction strategy can be formulated for the gradual introduction of DLTs. Additionally, blockchainification suggests – at least in the first phases of an introduction strategy – an emphasis on solutions where a DLT essentially just „replaces” the current information system support of already-digitized cooperations. While in the enterprise and industrial sphere blockchainification is already facilitated by an example-based understanding of key value dimensions, blockchain capabilities and value driver applications, for many other domains, these prerequisites are missing. Importantly, what is already available is not readily applicable for organizations involved in security activities in the broad sense; in many aspects, the value these organizations seek from IT systems is markedly different from the enterprise world. Thus, the paper proposes an initial key value dimension and supporting blockchain capability model for organizations involved in providing a select set of security services.