scholarly journals Customizable Vector Acceleration in Extreme-Edge Computing: A RISC-V Software/Hardware Architecture Study on VGG-16 Implementation

Electronics ◽  
2021 ◽  
Vol 10 (4) ◽  
pp. 518
Author(s):  
Stefano Sordillo ◽  
Abdallah Cheikh ◽  
Antonio Mastrandrea ◽  
Francesco Menichelli ◽  
Mauro Olivieri
Keyword(s):  
Power Efficiency ◽  
High Performance ◽  
Hardware Acceleration ◽  
The Internet ◽  
Classification Algorithms ◽  
Comprehensive Investigation ◽  
Convolutional Networks ◽  
Performance Requirements ◽  
Extreme Edge ◽  
The Internet Of Things

Computing in the cloud-edge continuum, as opposed to cloud computing, relies on high performance processing on the extreme edge of the Internet of Things (IoT) hierarchy. Hardware acceleration is a mandatory solution to achieve the performance requirements, yet it can be tightly tied to particular computation kernels, even within the same application. Vector-oriented hardware acceleration has gained renewed interest to support artificial intelligence (AI) applications like convolutional networks or classification algorithms. We present a comprehensive investigation of the performance and power efficiency achievable by configurable vector acceleration subsystems, obtaining evidence of both the high potential of the proposed microarchitecture and the advantage of hardware customization in total transparency to the software program.

scholarly journals Customizable Vector Acceleration in Extreme-Edge Computing: A RISC-V Software/Hardware Architecture Study on VGG-16 Implementation

2021 ◽  
Author(s):  
Mauro Olivieri ◽  
Abdallah Cheikh ◽  
Francesco Menichelli ◽  
Antonio Mastrandrea ◽  
Stefano Sordillo
Keyword(s):  
Cloud Computing ◽  
Power Efficiency ◽  
High Performance ◽  
Hardware Acceleration ◽  
High Potential ◽  
Classification Algorithms ◽  
Comprehensive Investigation ◽  
Convolutional Networks ◽  
Performance Requirements ◽  
Extreme Edge

Computing in the cloud-edge continuum, as opposed to cloud computing, relies on high performance processing on the extreme edge of the IoT hierarchy. Hardware acceleration is a mandatory solution to achieve the performance requirements, yet it can be tightly tied to particular computation kernels, even within the same application. Vector-oriented hardware acceleration has gained renewed interest to support AI applications like convolutional networks or classification algorithms. We present a comprehensive investigation of the performance and power efficiency achievable by configurable vector acceleration subsystems, obtaining evidence of both the high potential of the proposed microarchitecture and the advantage of hardware customization in total transparency to the software program.

Maximizing piezoelectricity by self-assembled highly porous perovskite–polymer composite films to enable the internet of things

2020 ◽  
Vol 8 (27) ◽  
pp. 13619-13629 ◽  
Author(s):  
Asif Abdullah Khan ◽  
Md Masud Rana ◽  
Guangguang Huang ◽  
Nanqin Mei ◽  
Resul Saritas ◽  
...  
Keyword(s):  
Internet Of Things ◽  
Polymer Composite ◽  
High Performance ◽  
Composite Films ◽  
The Internet ◽  
Next Generation ◽  
Self Assembled ◽  
Self Powered ◽  
The Internet Of Things ◽  
Highly Porous

A high-performance perovskite/polymer piezoelectric nanogenerator for next generation self-powered wireless micro/nanodevices.

Challenges and prospects of 3D micro-supercapacitors for powering the internet of things

2019 ◽  
Vol 12 (1) ◽  
pp. 96-115 ◽  
Author(s):  
Christophe Lethien ◽  
Jean Le Bideau ◽  
Thierry Brousse
Keyword(s):  
Energy Storage ◽  
Solid State ◽  
Internet Of Things ◽  
High Performance ◽  
Storage Systems ◽  
Electronic Devices ◽  
Electrochemical Energy Storage ◽  
The Internet ◽  
The Internet Of Things ◽  
Internet Of Thing

The fabrication of miniaturized electrochemical energy storage systems is essential for the development of future electronic devices for Internet of Thing applications. This paper aims at reviewing the current micro-supercapacitor technologies and at defining the guidelines to produce high performance micro-devices with special focuses onto the 3D designs as well as the fabrication of solid state miniaturized devices to solve the packaging issue.

Towards Using Cloud Elasticity on the Internet of Things Landscape

2019 ◽  
Vol 3 (1) ◽  
Author(s):  
Rodrigo Da Rosa Righi ◽  
Márcio Miguel Gomes ◽  
Cristiano Andrá Da Costa ◽  
Helge Parzyjegla ◽  
Hans-Ulrich Heiss
Keyword(s):  
Internet Of Things ◽  
Radio Frequency Identification ◽  
High Performance ◽  
Virtual Machines ◽  
The Internet ◽  
Data Volume ◽  
Significant Performance ◽  
Cloud Elasticity ◽  
Performance Gains ◽  
The Internet Of Things

The digital universe is growing at significant rates in recent years. One of the main responsible for this sentence is the Internet of Things, or IoT, which requires a middleware that should be capable to handle this increase of data volume at real-time. Particularly, data can arrive in the middleware in parallel as in terms of input data from Radio-Frequency Identification (RFID) readers as request-reply query operations from the users side. Solutions modeled at software, hardware and/or architecture levels present limitations to handle such load, facing the problem of scalability in the IoT scope. In this context, this arti- cle presents a model denoted Eliot - Elasticity-driven Internet of Things - which combines both cloud and high performance computing to address the IoT scal- ability problem in a novel EPCglobal-compliant architecture. Particularly, we keep the same API but offer an elastic EPCIS component in the cloud, which is designed as a collection of virtual machines (VMs) that are allocated and deallocated on-the-fly in accordance with the system load. Based on the Eliot model, we developed a prototype that could run over any black-box EPCglobal- compliant middleware. We selected the Fosstrak for this role, which is currently one of the most used IoT middlewares. Thus, the prototype acts as an upper layer over the Fosstrak to offer a better throughput and latency performances in an effortless way. The results are encouraging, presenting significant performance gains in terms of response time and request throughput when comparing both elastic (Eliot) and non-elastic (standard Fosstrak) executions.  

scholarly journals Self-switching diodes as RF rectifiers: evaluation methods and current progress

2019 ◽  
Vol 8 (2) ◽  
pp. 396-404 ◽  
Author(s):  
Nor Farhani Zakaria ◽  
Shahrir Rizal Kasjoo ◽  
Muammar Mohamad Isa ◽  
Zarimawaty Zailan ◽  
Mohd Khairuddin Md Arshad ◽  
...  
Keyword(s):  
Internet Of Things ◽  
High Frequency ◽  
Power Efficiency ◽  
Wireless Transmission ◽  
The Internet ◽  
Power Devices ◽  
New Paradigm ◽  
Iot Applications ◽  
Design Simplicity ◽  
The Internet Of Things

In the advancement of the Internet of Things (IoT) applications, widespread uses and applications of devices require higher frequency connectivity to be explored and exploited. Furthermore, the size, weight, power and cost demands for the IoT ecosystems also creates a new paradigm for the hardware where improved power efficiency and efficient wireless transmission needed to be investigated and made feasible. As such, functional microwave detectors to detect and rectify the signals transmitted in higher frequency regions are crucial. This paper reviewed the practicability of self switching diodes as Radio Frequency (RF) rectifiers. The existing methods used in the evaluation of the rectification performance and cut-off frequency are reviewed, and current achievements are then concluded. The works reviewed in this paper highlights the functionality of SSD as a RF rectifier with design simplicity, which may offer cheaper alternatives in current high frequency rectifying devices for application in low-power devices.

scholarly journals A High-Performance Coniform Helmholtz Resonator-Based Triboelectric Nanogenerator for Acoustic Energy Harvesting

Nanomaterials ◽  
2021 ◽  
Vol 11 (12) ◽  
pp. 3431
Author(s):  
Haichao Yuan ◽  
Hongyong Yu ◽  
Xiangyu Liu ◽  
Hongfa Zhao ◽  
Yiping Zhang ◽  
...  
Keyword(s):  
Energy Harvesting ◽  
Internet Of Things ◽  
High Performance ◽  
Helmholtz Resonator ◽  
Acoustic Energy ◽  
The Internet ◽  
Triboelectric Nanogenerator ◽  
Sensor Devices ◽  
Acoustic Energy Harvesting ◽  
The Internet Of Things

Harvesting acoustic energy in the environment and converting it into electricity can provide essential ideas for self-powering the widely distributed sensor devices in the age of the Internet of Things. In this study, we propose a low-cost, easily fabricated and high-performance coniform Helmholtz resonator-based Triboelectric Nanogenerator (CHR-TENG) with the purpose of acoustic energy harvesting. Output performances of the CHR-TENG with varied geometrical sizes were systematically investigated under different acoustic energy conditions. Remarkably, the CHR-TENG could achieve a 58.2% higher power density per unit of sound pressure of acoustic energy harvesting compared with the ever-reported best result. In addition, the reported CHR-TENG was demonstrated by charging a 1000 μF capacitor up to 3 V in 165 s, powering a sensor for continuous temperature and humidity monitoring and lighting up as many as five 0.5 W commercial LED bulbs for acoustic energy harvesting. With a collection features of high output performance, lightweight, wide frequency response band and environmental friendliness, the cleverly designed CHR-TENG represents a practicable acoustic energy harvesting approach for powering sensor devices in the age of the Internet of Things.

scholarly journals Hierarchically Authorized Transactions for Massive Internet-of-Things Data Sharing Based on Multilayer Blockchain

2019 ◽  
Vol 9 (23) ◽  
pp. 5159 ◽  
Author(s):  
Shichang Xuan ◽  
Yibo Zhang ◽  
Hao Tang ◽  
Ilyong Chung ◽  
Wei Wang ◽  
...  
Keyword(s):  
Cloud Computing ◽  
Big Data ◽  
Internet Of Things ◽  
Data Sharing ◽  
Privacy Protection ◽  
High Performance ◽  
The Internet ◽  
Blockchain Technology ◽  
And Storage ◽  
The Internet Of Things

With the arrival of the Internet of Things (IoT) era and the rise of Big Data, cloud computing, and similar technologies, data resources are becoming increasingly valuable. Organizations and users can perform all kinds of processing and analysis on the basis of massive IoT data, thus adding to their value. However, this is based on data-sharing transactions, and most existing work focuses on one aspect of data transactions, such as convenience, privacy protection, and auditing. In this paper, a data-sharing-transaction application based on blockchain technology is proposed, which comprehensively considers various types of performance, provides an efficient consistency mechanism, improves transaction verification, realizes high-performance concurrency, and has tamperproof functions. Experiments were designed to analyze the functions and storage of the proposed system.

scholarly journals PRIPRO: A Comparison of Classification Algorithms for Managing Receiving Notifications in Smart Environments

2020 ◽  
Vol 10 (2) ◽  
pp. 502 ◽  
Author(s):  
João Antônio Martins ◽  
Iago Sestrem Ochôa ◽  
Luis Augusto Silva ◽  
André Sales Mendes ◽  
Gabriel Villarrubia González ◽  
...  
Keyword(s):  
The Other ◽  
Statistical Hypothesis ◽  
The Internet ◽  
Smart Environments ◽  
Classification Algorithms ◽  
Intelligent Environments ◽  
Central Concept ◽  
Evolution Of Technology ◽  
To Receive ◽  
The Internet Of Things

With the evolution of technology over the years, it has become possible to develop intelligent environments based on the concept of the Internet of Things, distributed systems, and machine learning. Such environments are infused with various solutions to solve user demands from services. One of these solutions is the Ubiquitous Privacy (UBIPRI) middleware, whose central concept is to maintain privacy in smart environments and to receive notifications as one of its services. However, this service is freely performed, disregarding the privacy that the environment employs. Another consideration is that, based on the researched related work, it was possible to identify that the authors do not use statistical hypothesis tests in their solutions developed in the presented context. This work proposes an architecture for notification management in smart environments, composed by a notification manager named Privacy Notification Manager (PRINM) to assign it to UBIPRI and to aim to perform experiments between classification algorithms to delimit which one is most feasible to implement in the PRINM decision-making mechanism. The experiments showed that the J48 algorithm obtained the best results compared to the other algorithms tested and compared.

Sign in / Sign up

Export Citation Format

Share Document