In this paper, we propose a symbolic framework to analyze and debug communicating distributed models. We implement dedicated symbolic execution techniques for such models and use them to compute interaction scenarios ...
详细信息
Underground wireless sensors (UWASNs) are intended to serve a range of needs for global research and marine management. For some of these uses, the network of several sensors is positioned underwater at various depths...
详细信息
I/O is emerging as a major bottleneck for machine learning training, especially in distributed environments. Indeed, at large scale, I/O takes as much as 857, of training time. Addressing this I/O bottleneck necessity...
详细信息
ISBN:
(纸本)9781450384421
I/O is emerging as a major bottleneck for machine learning training, especially in distributed environments. Indeed, at large scale, I/O takes as much as 857, of training time. Addressing this I/O bottleneck necessity les careful optimization, as optimal data ingestion pipelines differ between systems, and require a delicate balance between access to local storage, external filesysleins, and remote nodes. We introduce NoPFS, a machine learning I/O middleware, which provides a scalable, flexible, and easy-to -use solution to the I/O bottleneck. NoPFS uses clairvoyance: Given the seed generating the random access pattern for training with SG1), it can exactly predict when and where a sample will be accessed. We combine this with an analysis of access patterns and a performance model to provide distributed caching policies that adapt to different datasets and storage hierarchies. NoPFS reduces I/O times and improves end to -end training by up to 5.4x on the ImageNet-lk;ImageNet-22k;and CasinoHow datasets.
There is an increasing demand for unmanned systems to perform a wide range of Intelligence, Surveillance and Reconnaissance (ISR) missions. The demand for small unmanned air systems (SUAS) results from reducing costs,...
详细信息
ISBN:
(纸本)9781728174365
There is an increasing demand for unmanned systems to perform a wide range of Intelligence, Surveillance and Reconnaissance (ISR) missions. The demand for small unmanned air systems (SUAS) results from reducing costs, increasing use for stand-off monitoring, and replacing piloted aircraft in disaster responses. With reduced size come more restrictive payload weights, limiting the number and diversity of sensors that can be located on a single platform. As a result, platforms must coordinate to share and leverage multi-sensor data over limited communications channels. This paper presents an on-device design and implementation of a distributed joint manifold learning (DJML) approach for improved object detection, classification, and identification (DCI). The DJML design addresses the joint utilization of sensor data from a collection of decentralized, heterogeneous sensing platforms (e.g., sensor swarm) in dynamic environments with constrained communications. This paper focuses on the implementation and evaluation of the hardware tradeoffs for distributed device mechanisms: drone-carried sensing, communication, and computing. Different sensor modality data are preloaded on SD (secure digital) cards and sequentially processed on local devices to emulate the additional sensor modalities. The in-lab testing results demonstrate the robustness and resiliency of on-device decentralized DCI under various conditions of sensor placement and communication degradation between platforms.
The ability of IoT devices to sense and share various physical parameters plays a key role in a smart system. Along side benefits, it also bears the potential to cause a breach of privacy for the users of the smart-sy...
详细信息
ISBN:
(数字)9798350381993
ISBN:
(纸本)9798350382006
The ability of IoT devices to sense and share various physical parameters plays a key role in a smart system. Along side benefits, it also bears the potential to cause a breach of privacy for the users of the smart-systems. Existing solutions for privacy-preserving computation for distributedsystems either extensively use highly complex cryptographic techniques or exploit an extremely high degree of message passing among the devices through secure channels. However, for the resource-constrained IoT devices, which compose a significant fraction of the smart-systems, the existing solutions for privacy-preserving computation strategies does not fulfill the current requirements. To address this issue, in this work, we propose a novel real-time lightweight strategy LiPI for Privacy-Preserving Data Aggregation in low-power IoT systems. LiPI uses lightweight distributed and collaborative data obfuscation, which substantially minimizes the computation requirements. In addition, it exploits the recent advances in Synchronous-Transmission (st)-based protocols to efficiently fulfill the communication requirements too, making it efficient to work in real-time. Furthermore, LiPI also avoids dependency on any trusted third party. Extensive evaluation based on comprehensive experiments in both simulation platforms and publicly available WSN/IoT testbeds demonstrates that our strategy achieves the goal at least 51.7% faster and consumes 50.5% lesser energy compared to the existing state-of-the-art strategies.
In conjunction with the Internet of Things (IoT), the rapid development of blockchain technology is bringing about a revolution in the administration of assets, the provision of secure online communication, and the pr...
详细信息
Clustering technique is one of the reliable methods to enhance the lifetime of the routed network in wireless sensor networks. The bunch nodes in the distributed network nominate cluster heads based on unanimous or el...
详细信息
The proceedings contain 44 papers. The special focus in this conference is on Advances in Signal Processing and Communication Engineering. The topics include: Single-Precision Floating-Point Multiplier Design Using Qu...
ISBN:
(纸本)9789811955495
The proceedings contain 44 papers. The special focus in this conference is on Advances in Signal Processing and Communication Engineering. The topics include: Single-Precision Floating-Point Multiplier Design Using Quantum-Dot Cellular Automata with Power Dissipation Analysis;compression Techniques for Low Power Hardware Accelerator Design: Case studies;sequence Set Design for Radar System;design of an All Digital Phase-Locked Loop Using Cordic Algorithm;analysis of Deep Learning Algorithms for Image Denoising;an Extensive Survey on Assessment of Multicore Processors for Embedded systems;image Segmentation Techniques and Optimization Algorithms for Lung Cancer Detection;handwritten to Text Document Converter;An Efficient Energy Aware for Reliable Route Discovery Using Energy with Movement Detection Technique in MANET;Lumped Circuit Modeling at Nanoscale (Part-II: Coupling Between Two Nanospheres);design and Implementation of Imprecise Adders for Low-Power Applications;speech Processed Public Addressing System;an Approach Towards Data Privacy Issues in distributed Cyber Physical System;Classification of LPI Radar Signals Using Multilayer Perceptron (MLP) Neural Networks;a Systematic Review on Screening of Diabetic Retinopathy and Maculopathy Using Artificial Intelligence;Area Efficient and High-Throughput Radix-4 1024-Point FFT Processor for DSP Applications;air Quality Monitoring System Based on Artificial Intelligence;an Improved Technique for Identification of Forgery Image Detection Using Clustering Method;EEG Signal Analysis During stroop Task for Checking the Effect of Sleep Deprivation;UWB Localization Procedures with Range Control Methods—A Review;deep Learning Model for Multiclass Classification of Diabetic Retinal Fundus Images Using Gradient Descent Optimization;multilevel Authentication to Wireless sensor Networks Against Malicious Attacks Using Butterfly Method.
Integrating distributed Generation (DG) sources and Electric Vehicles (EVs) into radial distribution systems (RDS) is a promising approach to enhancing power system efficiency and reliability. This trend is driven by ...
详细信息
ISBN:
(数字)9798350384246
ISBN:
(纸本)9798350384253
Integrating distributed Generation (DG) sources and Electric Vehicles (EVs) into radial distribution systems (RDS) is a promising approach to enhancing power system efficiency and reliability. This trend is driven by the deregulation of the electric power sector and technical constraints in extending distribution and transmission networks to some areas. By carefully selecting the optimal location and sizing for DG sources and electric vehicles (EVs), it is possible to minimize system losses, improve voltage profiles, and enhance overall reliability. This study suggests the integration of multiple DGs and EV s into the system and how the PLoss get affected along with voltage profiles. However, the PLoss increase if the number of DGs exceeds their optimal level. This study integrates 3 DGs and 3 EVs into the RDS separately. The location of the DGs and EVs are obtained with the help of the voltage stability index, and sizes were obtained using the Whale Optimization Algorithm (WOA). It is implemented into the ieee 33 test system to verify its robustness and effectiveness. The results show a loss reduction of 41.96%, 57.52%, and 64.73% for the ieee bus system with the integration of one, two and three DGs, respectively and a slight increase in power loss concerning EV integration as the system load increases. In addition, the voltage stability index is improved significantly for DGs but was affected in the case of EVs.
Currently, integrating battery energy storage system (BESS) with the renewable energy resources is one of the potential ways to maintain the quality and reliability of the power system. This paper investigates the rec...
详细信息
ISBN:
(数字)9798350384246
ISBN:
(纸本)9798350384253
Currently, integrating battery energy storage system (BESS) with the renewable energy resources is one of the potential ways to maintain the quality and reliability of the power system. This paper investigates the recent advancements and challenges in grid connected BESS. Short overviews of the mechanical, electrical, electrochemical, chemical, and thermal BESS technologies are provided. The presentation includes a review of typical power converter topologies, such as transformer-based, transformer less with distributed or common dc-link, and hybrid systems, as well as a few observations on how to integrate advanced grid support features into the BESS control. This article explores new technologies, such as flexible photovoltaic system power regulation, hydrogen, and their applications in second-life electric vehicle (EV) batteries. Finally challenges in implementing and managing grid connected BESSs are explored are discussed.
暂无评论