Search results for "embedded"
showing 10 items of 412 documents
Investigating the Impact of Radiation-Induced Soft Errors on the Reliability of Approximate Computing Systems
2020
International audience; Approximate Computing (AxC) is a well-known paradigm able to reduce the computational and power overheads of a multitude of applications, at the cost of a decreased accuracy. Convolutional Neural Networks (CNNs) have proven to be particularly suited for AxC because of their inherent resilience to errors. However, the implementation of AxC techniques may affect the intrinsic resilience of the application to errors induced by Single Events in a harsh environment. This work introduces an experimental study of the impact of neutron irradiation on approximate computing techniques applied on the data representation of a CNN.
Highly Performant, Deep Neural Networks with sub-microsecond latency on FPGAs for Trigger Applications
2020
Artificial neural networks are becoming a standard tool for data analysis, but their potential remains yet to be widely used for hardware-level trigger applications. Nowadays, high-end FPGAs, often used in low-level hardware triggers, offer theoretically enough performance to include networks of considerable size. This makes it very promising and rewarding to optimize a neural network implementation for FPGAs in the trigger context. Here an optimized neural network implementation framework is presented, which typically reaches 90 to 100% computational efficiency, requires few extra FPGA resources for data flow and controlling, and allows latencies in the order of 10s to few 100s of nanoseco…
Neural Classification of HEP Experimental Data
2009
High Energy Physics (HEP) experiments require discrimination of a few interesting events among a huge number of background events generated during an experiment. Hierarchical triggering hardware architectures are needed to perform this tasks in real-time. In this paper three neural network models are studied as possible candidate for such systems. A modified Multi-Layer Perception (MLP) architecture and a E alpha Net architecture are compared against a traditional MLP Test error below 25% is archived by all architectures in two different simulation strategies. E alpha Net performance are 1 to 2% better on test error with respect to the other two architectures using the smaller network topol…
Support Tool for the Combined Software/Hardware Design of On-Chip ELM Training for SLFF Neural Networks
2016
Typically, hardware implemented neural networks are trained before implementation. Extreme learning machine (ELM) is a noniterative training method for single-layer feed-forward (SLFF) neural networks well suited for hardware implementation. It provides fixed-time learning and simplifies retraining of a neural network once implemented, which is very important in applications demanding on-chip training. This study proposes the data flow of a software support tool in the design process of a hardware implementation of on-chip ELM learning for SLFF neural networks. The software tool allows the user to obtain the optimal definition of functional and hardware parameters for any application, and e…
Fingerprint Quality Evaluation in a Novel Embedded Authentication System for Mobile Users
2015
The way people access resources, data and services, is radically changing using modern mobile technologies. In this scenario, biometry is a good solution for security issues even if its performance is influenced by the acquired data quality. In this paper, a novel embedded automatic fingerprint authentication system (AFAS) for mobile users is described. The goal of the proposed system is to improve the performance of a standard embedded AFAS in order to enable its employment in mobile devices architectures. The system is focused on the quality evaluation of the raw acquired fingerprint, identifying areas of poor quality. Using this approach, no image enhancement process is needed after the …
Fluorescence In Situ Hybridization (FISH) on Formalin-Fixed Paraffin-Embedded (FFPE) Tissue Sections
2011
Fluorescence In Situ Hybridization (FISH) is a powerful technique for localizing specific DNA targets directly in the fixed tissue or cells. Bacterial artificial chromosome (BAC) as well as commercial probes, which could be supplied ready for use or concentrated and must be diluted following the manufacturers instructions, can be used. The technique requires 2 days, as an overnight incubation of the FISH probes is needed for optimal hybridization. The critical steps include deparaffinization of tissue sections, optimal pretreatment (target retrieval and protein digestion), and probe hybridization. In this chapter, the described FISH protocol provides a methodology for analyzing the cytogene…
On-board Energy Consumption Assessment for Symbolic Execution Models on Embedded Devices
2020
Internet of Things (IoT) applications operate in several domains while requiring seamless integration among heterogeneous objects. Regardless of the specific platform and context, IoT applications demand high energy efficiency. Adopting resource-constrained embedded devices for IoT applications means ensuring low power consumption, low maintenance costs and possibly longer battery life. Meeting these requirements is particularly arduous as programmers are not able to monitor the energy consumption of their own software during development or when applications are finally deployed. In this paper, we discuss on-board real-time energy evaluation of both hardware and software during the developm…
Solar Inexhaustible Power Source for Wireless Sensor Node
2008
Currently the appearance of really low power wireless transceivers is motivating the use of renewable energies to power embedded wireless sensor nodes in many applications. Nevertheless, energy storage and its degradation still keep on being the main issues in the design of any battery powered device. We present an autonomous power source based on a new system architecture, which uses the energy scavenging to replenish two different rechargeable energy buffers instead of the conventional single one. Combining appropriately a degradable large backup battery (Lithium-Ion) and a shorter but non degradable storage element (Supercapacitor), the lifetime of the group can be widely extended to wha…
Adapting power consumption to performance requirements in a MSP430 microcontroller
2005
Microcontroller based low power systems using batteries must run at low clock frequencies with short activity periods to extend battery life. These constraints reduce the microcontroller computational capabilities and limit the complexity of algorithms that can be used. New microcontroller architectures, as the Texas Instruments MSP430 family, allow to adapt power consumption to application performance requirements combining several low power modes with the capability of switching the clock frequency dynamically. A complete characterization of power consumption vs. performance has been obtained for the MSP430. A method to determine the range of clock frequencies that meets a consumption and…
Real-time Sound Source Localization on an Embedded GPU Using a Spherical Microphone Array
2015
Abstract Spherical microphone arrays are becoming increasingly important in acoustic signal processing systems for their applications in sound field analysis, beamforming, spatial audio, etc. The positioning of target and interfering sound sources is a crucial step in many of the above applications. Therefore, 3D sound source localization is a highly relevant topic in the acoustic signal processing field. However, spherical microphone arrays are usually composed of many microphones and running signal processing localization methods in real time is an important issue. Some works have already shown the potential of Graphic Processing Units (GPUs) for developing high-end real-time signal proce…