Showing papers on "Systems architecture published in 2017"

A Manufacturing Big Data Solution for Active Preventive Maintenance

Abstract: Industry 4.0 has become more popular due to recent developments in cyber-physical systems, big data, cloud computing, and industrial wireless networks. Intelligent manufacturing has produced a revolutionary change, and evolving applications, such as product lifecycle management, are becoming a reality. In this paper, we propose and implement a manufacturing big data solution for active preventive maintenance in manufacturing environments. First, we provide the system architecture that is used for active preventive maintenance. Then, we analyze the method used for collection of manufacturing big data according to the data characteristics. Subsequently, we perform data processing in the cloud, including the cloud layer architecture, the real-time active maintenance mechanism, and the offline prediction and analysis method. Finally, we analyze a prototype platform and implement experiments to compare the traditionally used method with the proposed active preventive maintenance method. The manufacturing big data method used for active preventive maintenance has the potential to accelerate implementation of Industry 4.0.

ScaleDeep: A Scalable Compute Architecture for Learning and Evaluating Deep Networks

Abstract: Deep Neural Networks (DNNs) have demonstrated state-of-the-art performance on a broad range of tasks involving natural language, speech, image, and video processing, and are deployed in many real world applications. However, DNNs impose significant computational challenges owing to the complexity of the networks and the amount of data they process, both of which are projected to grow in the future. To improve the efficiency of DNNs, we propose ScaleDeep, a dense, scalable server architecture, whose processing, memory and interconnect subsystems are specialized to leverage the compute and communication characteristics of DNNs. While several DNN accelerator designs have been proposed in recent years, the key difference is that ScaleDeep primarily targets DNN training, as opposed to only inference or evaluation. The key architectural features from which ScaleDeep derives its efficiency are: (i) heterogeneous processing tiles and chips to match the wide diversity in computational characteristics (FLOPs and Bytes/FLOP ratio) that manifest at different levels of granularity in DNNs, (ii) a memory hierarchy and 3-tiered interconnect topology that is suited to the memory access and communication patterns in DNNs, (iii) a low-overhead synchronization mechanism based on hardware data-flow trackers, and (iv) methods to map DNNs to the proposed architecture that minimize data movement and improve core utilization through nested pipelining. We have developed a compiler to allow any DNN topology to be programmed onto ScaleDeep, and a detailed architectural simulator to estimate performance and energy. The simulator incorporates timing and power models of ScaleDeep's components based on synthesis to Intel's 14nm technology. We evaluate an embodiment of ScaleDeep with 7032 processing tiles that operates at 600 MHz and has a peak performance of 680 TFLOPs (single precision) and 1.35 PFLOPs (half-precision) at 1.4KW. Across 11 state-of-the-art DNNs containing 0.65M-14.9M neurons and 6.8M-145.9M weights, including winners from 5 years of the ImageNet competition, ScaleDeep demonstrates 6x-28x speedup at iso-power over the state-of-the-art performance on GPUs.

Ultra Low Power Wake-Up Radios: A Hardware and Networking Survey

Abstract: In wireless environments, transmission and reception costs dominate system power consumption, motivating research effort on new technologies capable of reducing the footprint of the radio, paving the way for the Internet of Things. The most important challenge is to reduce power consumption when receivers are idle, the so called idle-listening cost. One approach proposes switching off the main receiver, then introduces new wake-up circuitry capable of detecting an incoming transmission, optionally discriminating the packet destination using addressing, then switching on the main radio only when required. This wake-up receiver technology represents the ultimate frontier in low power radio communication. In this paper, we present a comprehensive literature review of the research progress in wake-up radio (WuR) hardware and relevant networking software. First, we present an overview of the WuR system architecture, including challenges to hardware design and a comparison of solutions presented throughout the last decade. Next, we present various medium access control and routing protocols as well as diverse ways to exploit WuRs, both as an extension of pre-existing systems and as a new concept to manage low-power networking.

5G Cellular User Equipment: From Theory to Practical Hardware Design

Abstract: Research and development on the next generation wireless systems, namely 5G, has experienced explosive growth in recent years. In the physical layer, the massive multiple-input-multiple-output (MIMO) technique and the use of high GHz frequency bands are two promising trends for adoption. Millimeter-wave (mmWave) bands, such as 28, 38, 64, and 71 GHz, which were previously considered not suitable for commercial cellular networks, will play an important role in 5G. Currently, most 5G research deals with the algorithms and implementations of modulation and coding schemes, new spatial signal processing technologies, new spectrum opportunities, channel modeling, 5G proof of concept systems, and other system-level enabling technologies. In this paper, we first investigate the contemporary wireless user equipment (UE) hardware design, and unveil the critical 5G UE hardware design constraints on circuits and systems. On top of the said investigation and design tradeoff analysis, a new, highly reconfigurable system architecture for 5G cellular user equipment, namely distributed phased arrays based MIMO (DPA-MIMO) is proposed. Finally, the link budget calculation and data throughput numerical results are presented for the evaluation of the proposed architecture.

The Technology Behind Personal Digital Assistants: An overview of the system architecture and key components

Abstract: We have long envisioned that one day computers will understand natural language and anticipate what we need, when and where we need it, and proactively complete tasks on our behalf. As computers get smaller and more pervasive, how humans interact with them is becoming a crucial issue. Despite numerous attempts over the past 30 years to make language understanding (LU) an effective and robust natural user interface for computer interaction, success has been limited and scoped to applications that were not particularly central to everyday use. However, speech recognition and machine learning have continued to be refined, and structured data served by applications and content providers has emerged. These advances, along with increased computational power, have broadened the application of natural LU to a wide spectrum of everyday tasks that are central to a user's productivity. We believe that as computers become smaller and more ubiquitous [e.g., wearables and Internet of Things (IoT)], and the number of applications increases, both system-initiated and user-initiated task completion across various applications and web services will become indispensable for personal life management and work productivity. In this article, we give an overview of personal digital assistants (PDAs); describe the system architecture, key components, and technology behind them; and discuss their future potential to fully redefine human?computer interaction.

Integrated photovoltaic-grid dc fast charging system for electric vehicle: A review of the architecture and control

Abstract: The fast charger for electric vehicle (EV) is a complex system that incorporates numerous interconnected subsystems. The interactions among these subsystems require a holistic understanding of the system architecture, control, power electronics, and their overall interaction with the electrical grid system. This review paper presents important aspects of a PV-grid integrated dc fast charger—with a special focus on the charging system components, architecture, operational modes, and control. These include the interaction between the PV power source, grid electricity, energy storage unit (ESU) and power electronics for the chargers. A considerable amount of discussion is also dedicated to battery management systems (BMS) and their mutual interactions in the control processes. For the power electronics, the paper evaluates soft switching non-isolated dc-to-dc power converter topologies that can be possibly used as future EV chargers. In addition to these, a brief discussion on the impact of the PV-grid charging on the ac grid and distribution system and their remedial measures are presented. Furthermore, the challenges in regard to the vehicle to grid (V2G) concept are also described. It is envisaged that the information provided in this paper would be useful as a one-stop document for engineers, researchers and others who require information related to the dc fast charging of EV that incorporates a renewable energy source.

RF-scanner: Shelf scanning with robot-assisted RFID systems

Abstract: Shelf scanning is one of the most important processes for inventory management in a library. It helps the librarians and library users discover the miss-shelved books and pinpoint where they are, improving the quality of service. By traditional means, however, manually checking each bookshelf suffers from extremely intensive labor and long scanning delay. Although some existing RFID-enabled approaches have been proposed, they suffer from either high-cost infrastructure or complicated system deployment, forming a great barrier to commercial adoption. In light of this, we in this paper propose a smart system called RF-Scanner that can perform the shelf scanning automatically by combining the robot technology and the RFID technology. The former is used for replacing the librarians and liberating them from intensively manual labor. The later is installed on the robot and moves with the robot to scan the on-the-shelf books. We formulate two important issues concerned by librarians, the book localization and the lying-down book detection, and give the sophisticated solutions to them. Besides, we implement RF-Scanner and put it into practical use in our school library. Long-term experiments and studies show that RF-Scanner provides fine-grained book localization with a mean error of just 1.3 cm and accurate detection accuracy of lying-down books with a mean error of 6%.

Multi-agent systems and their applications

Abstract: The number of distributed energy components and devices continues to increase globally. As a result, distributed control schemes are desirable for managing and utilizing these devices, together wit...

maplab: An Open Framework for Research in Visual-inertial Mapping and Localization

Abstract: Robust and accurate visual-inertial estimation is crucial to many of today's challenges in robotics. Being able to localize against a prior map and obtain accurate and driftfree pose estimates can push the applicability of such systems even further. Most of the currently available solutions, however, either focus on a single session use-case, lack localization capabilities or an end-to-end pipeline. We believe that only a complete system, combining state-of-the-art algorithms, scalable multi-session mapping tools, and a flexible user interface, can become an efficient research platform. We therefore present maplab, an open, research-oriented visual-inertial mapping framework for processing and manipulating multi-session maps, written in C++. On the one hand, maplab can be seen as a ready-to-use visual-inertial mapping and localization system. On the other hand, maplab provides the research community with a collection of multisession mapping tools that include map merging, visual-inertial batch optimization, and loop closure. Furthermore, it includes an online frontend that can create visual-inertial maps and also track a global drift-free pose within a localization map. In this paper, we present the system architecture, five use-cases, and evaluations of the system on public datasets. The source code of maplab is freely available for the benefit of the robotics research community.

System-Level Design Considerations for Digital Pre-Distortion of Wireless Base Station Transmitters

Abstract: Over the past 25 years or so there has been much interest in the use of digital pre-distortion (DPD) techniques for the linearization of RF and microwave power amplifiers. In this paper, we describe the important system and hardware requirements for the four main subsystems found in the DPD linearized transmitter: RF/analog, data converters, digital signal processing, and the DPD architecture and algorithms, and illustrate how the overall DPD system architecture is influenced by the design choices that may be made in each of these subsystems. We shall also consider the challenges presented to future applications of DPD systems for wireless communications, such as higher operating frequencies, wider signal bandwidths, greater spectral efficiency signals, resulting in higher peak-to-average power ratios, multiband and multimode operation, lower power consumption requirements, faster adaption, and how these affect the system design choices.

Monotasks: Architecting for Performance Clarity in Data Analytics Frameworks

Abstract: In today's data analytics frameworks, many users struggle to reason about the performance of their workloads. Without an understanding of what factors are most important to performance, users can't determine what configuration parameters to set and what hardware to use to optimize runtime. This paper explores a system architecture designed to make it easy for users to reason about performance bottlenecks. Rather than breaking jobs into tasks that pipeline many resources, as in today's frameworks, we propose breaking jobs into monotasks: units of work that each use a single resource. We demonstrate that explicitly separating the use of different resources simplifies reasoning about performance without sacrificing performance. Monotasks provide job completion times within 9% of Apache Spark for typical scenarios, and lead to a model for job completion time that predicts runtime under different hardware and software configurations with at most 28% error. Furthermore, separating the use of different resources allows for new optimizations to improve performance.

Workload management for dynamic mobile device clusters in edge femtoclouds

Abstract: Edge computing offers an alternative to centralized, in-the-cloud compute services. Among the potential advantages of edge computing are lower latency that improves responsiveness, reduced wide-area network congestion, and possibly greater privacy by keeping data more local. In our previous work on Femtoclouds, we proposed taking advantage of clusters of devices that tend to be co-located in places such as public transit, classrooms or coffee shops. These clusters can perform computations for jobs generated from within or outside of the cluster. In this paper, we address the full requirements of workload management in Femtoclouds. These functions enable a Femtocloud to provide a service to job initiators that is similar to that provided by a centralized cloud service. We develop a system architecture that relies on the cloud to efficiently control and manage a Femtocloud. Within this architecture, we develop adaptive workload management mechanisms and algorithms to manage resources and effectively mask churn. We implement a prototype of our Femtocloud system on Android devices and utilize it to evaluate the overall system performance. We use simulation to isolate and study the impact of our workload management mechanisms and test the system at scale. Our prototype and simulation results demonstrate the efficiency of the Femtocloud workload management mechanisms especially in situations with potentially high churn. For instance, our mechanisms can reduce the average job completion time by up to 26% compared to similar mechanisms used in traditional cloud computing systems when used in situations that suggest high churn.

Real-time wireless multisensory smart surveillance with 3D-HEVC streams for internet-of-things (IoT)

Abstract: This paper presents the design of a novel, real-time, wireless, multisensory, smart surveillance system with 3D-HEVC features. The proposed high-level system architecture of the surveillance system is analyzed. The advantages of HEVC encoding are presented. Methods for synchronization between multiple streams are presented. Available wireless standards are presented and compared. A network-adaptive transmission protocol for a reliable, real-time, multisensory surveillance system is proposed. Adaptive packet frame grouping (APFG) and adaptive quantization are deployed to maximize the quality-of-experience (QoE). Measurements of the proposed protocol have been shown to provide superior results compared to existing transport protocols.

An Integrated System for Perception-Driven Autonomy with Modular Robots.

Abstract: The theoretical ability of modular robots to reconfigure in response to complex tasks in a priori unknown environments has frequently been cited as an advantage and remains a major motivator for work in the field. We present a modular robot system capable of autonomously completing high-level tasks by reactively reconfiguring to meet the needs of a perceived, a priori unknown environment. The system integrates perception, high-level planning, and modular hardware, and is validated in three hardware demonstrations. Given a high-level task specification, a modular robot autonomously explores an unknown environment, decides when and how to reconfigure, and manipulates objects to complete its task. The system architecture balances distributed mechanical elements with centralized perception, planning, and control. By providing an example of how a modular robot system can be designed to leverage reactive reconfigurability in unknown environments, we have begun to lay the groundwork for modular self-reconfigurable robots to address tasks in the real world.

Peculiarities of content forming and analysis in internet newspaper covering music news

Abstract: The actual task of research and development of methods and means for processing data in intellectual information systems forming content with the use of the classification, mathematical and software means and a generalized system architecture. The necessity for development of methods and means of processing data in intelligent information systems forming content by improving system architecture in order to automate processes of formation, management and marketing of content has been justified.

High-Performance Computing with Quantum Processing Units

Abstract: The prospects of quantum computing have driven efforts to realize fully functional quantum processing units (QPUs). Recent success in developing proof-of-principle QPUs has prompted the question of how to integrate these emerging processors into modern high-performance computing (HPC) systems. We examine how QPUs can be integrated into current and future HPC system architectures by accounting for functional and physical design requirements. We identify two integration pathways that are differentiated by infrastructure constraints on the QPU and the use cases expected for the HPC system. This includes a tight integration that assumes infrastructure bottlenecks can be overcome as well as a loose integration that assumes they cannot. We find that the performance of both approaches is likely to depend on the quantum interconnect that serves to entangle multiple QPUs. We also identify several challenges in assessing QPU performance for HPC, and we consider new metrics that capture the interplay between system architecture and the quantum parallelism underlying computational performance.

An improved Cyber-Physical Systems architecture for Industry 4.0 smart factories

Abstract: The Cyber-Physical System (CPS) is the core concept of Industry 4.0 for building smart factories. We can rely on the ISA-95 architecture or the 5C architecture to build the CPS. However, both architectures emphasize more on vertical integration, and less on horizontal integration. This paper proposes the 8C architecture by adding 3C facets into the 5C architecture. The 3C facets are coalition, customer, and content. The proposed 8C architecture is a helpful guideline for us to build the CPS for smart factories.

Supporting design via the System Operational Dependency Analysis methodology

Abstract: In this paper, we introduce the system operational dependency analysis methodology. Its purpose is to assess the effect of dependencies between components in a monolithic complex system, or between systems in a system-of-systems, and to support design decision making. We propose a parametric model of the behavior of the system. This approach results in a simple, intuitive model, whose parameters give a direct insight into the causes of observed, and possibly emergent, behavior. Using the proposed method, designers, and decision makers can quickly analyze and explore the behavior of complex systems and evaluate different architecture under various working conditions. Thus, the system operational dependency analysis method supports educated decision making both in the design and in the update process of systems architecture, without the need to execute extensive simulations. In particular, in the phase of concept generation and selection, the information given by the method can be used to identify promising architectures to be further tested and improved, while discarding architectures that do not show the required level of global features. Application of the proposed method to a small example is used to demonstrate both the validation of the parametric model, and the capabilities of the method for system analysis, design and architecture.

Automotive Electrical and Electronic Architecture Security via Distributed In-Vehicle Traffic Monitoring

Abstract: Due to the growing interconnectedness and complexity of in-vehicle networks, in addition to safety, security is becoming an increasingly important topic in the automotive domain. In this paper, we study techniques for detecting security infringements in automotive electrical and electronic (E/E) architectures. Toward this we propose in-vehicle network traffic monitoring to detect increased transmission rates of manipulated message streams. Attacks causing timing violations can disrupt safety-critical functions and have severe consequences. To reduce costs and prevent single points of failure, our approach enables an automatic distribution of detection tasks among selected E/E architecture components, such as a subset of electronic control units. First, we analyze a concrete E/E system architecture to determine the communication parameters and properties necessary for detecting security attacks. These are then used for a parametrization of the corresponding detection algorithms and the distribution of attack detection tasks. We use a lightweight message monitoring method and optimize the placement of detection tasks to ensure a full-coverage of the E/E system architecture and a timely detection of an attack.

Prototype of group heart rate monitoring with NODEMCU ESP8266

Abstract: Paper describes the development of prototype that enables monitoring of heart rate and inter beat interval for several subjects. The prototype was realized using ESP8266 hardware modules, WebSocket library, nodejs and JavaScript. System architecture is described where nodejs server acts as the signal processing and GUI code provider for clients. Signal processing algorithm was implemented in JavaScript. Application GUI is presented which can be used on mobile devices. Several important parts of the code are described which illustrate the communication between ESP8266 modules, server and clients. Developed prototype shows one of the possible realizations of group monitoring of biomedical data.

Probabilistic Threat Detection for Risk Management in Cyber-physical Medical Systems

Abstract: Medical devices are complex cyber-physical systems incorporating emergent hardware and software components. However, this complexity leads to a wide attack surface posing security risks and vulnerabilities. Mitigation and management of such risks during premarket design and postmarket deployment are required. Dynamically mitigating threat potential in the presence of unknown vulnerabilities requires an adaptive risk-based scheme to assess the system’s state, a secure system architecture that can isolate hardware and software components, and design methods that can adaptively adjust the system’s topology based on risk changes. The essential complementary aspects during deployment are detecting, characterizing, and quantifying security threats. This article presents a dynamic risk management and mitigation approach based on probabilistic threat estimation. A smart-connected-pacemaker case study illustrates the approach. This article is part of a special issue on Software Safety and Security Risk Mitigation in Cyber-physical Systems.