1,320 research outputs found
Further properties of Azimi-Hagler Banach spaces
No full textsummary:For the Azimi-Hagler spaces more geometric and topological properties are investigated. Any constructed space is denoted by . We show \item {(i)} The subspace generated by a subsequence of is complemented. \item {(ii)} The identity operator from to when is unbounded. \item {(iii)} Every bounded linear operator on some subspace of is compact. It is known that if any is a dual space, then \item {(iv)} duals of spaces contain isometric copies of and their preduals contain asymptotically isometric copies of . \item {(v)} We investigate the properties of the operators from spaces to their predual
Expression analysis of protein inhibitor of activated STAT (PIAS) genes in IFNβ-treated multiple sclerosis patients [Corrigendum]
No full textTaheri M, Azimi G, Sayad A, et al. J Inflamm Res. 2018;11:457–463.On page 457, Author list and Correspondence, the last author’s name was misspelt. The correct name is Soudeh Ghafouri-Fard.Read the original articl
Dataset for FII to Economic growth
No full textThis study employs a set of panel data from the World Bank's World Development Indicators (WDI), IMF's Financial Access Survey, and World Bank's Worldwide Governance Indicators to test the effects of financial inclusion on economic growth
CNN-Oriented Placement Algorithm for High-Performance Accelerators on Rad-Hard FPGAs
No full textConvolutional Neural Networks (CNNs) are quickly becoming one of the most common applications running on hardware accelerators. Considering Field Programmable Gate Arrays (FPGAs), due to their high flexibility and computational performance, they are suitable for fast classification tasks and therefore, pave the way for new machine learning inference approaches. In this work, we first designed a fully interconnected CNN architecture implementable on a single FPGA. Secondly, we developed a new Neural Node-oriented placement algorithm to enable resilient CNN accelerators on space-grade FPGAs. The proposed solution reduces the single event transient error sensitivity of CNN single neuron cores while achieving high performance and effective overall convolutional architecture fault tolerance. The developed approach has been applied and integrated into a state-of-the-art Radiation Tolerant FPGAs (RTG4) implementation flow. The experimental evaluation has been performed on a Microchip test board through benchmark application performance evaluation and transient error analysis. Experimental results demonstrate an improvement of 27.2% of the maximal working frequency and a reduction of the transient error sensitivity of about three times with respect to the previous mitigation approaches
On the evaluation of SEU effects on AXI interconnect within AP-SoCs
No full textG-Programmable System-on-Chips offering the union of a processor system with a programmable hardware gave rise to applications that choose hardware acceleration to offload and parallelize computationally demanding tasks. Due to flexibility and performance they provide at low cost, these devices are also appealing for several applications in avionics, aerospace and automotive sectors, where reliability is the main concern. In particular, the interconnection architecture, and especially the AXI Interconnection for FPGA-accelerated applications, plays a critical role in these systems. This paper presents a reliability analysis of the AXI Interconnect IP Core implemented on Zynq-7000 AP-SoC against SEUs in the configuration memory of the programmable logic. The analysis has been conducted performing a fault injection campaign on the specific section of the configuration memory implementing the IP Core under test, which has been implemented within a benchmark design. The results are analyzed and classified, highlighting the criticality of the AXI Interconnect IP Core as a point of failure, especially for SEU-hardened hardware accelerator relying on mitigation techniques based on fine-grained and coarse-grained replication
FireNN: Neural Networks Reliability Evaluation on Hybrid Platforms
No full textThe growth of neural networks complexity has led to adopt of hardware-accelerators to cope with the computational power required by the new architectures. The possibility to adapt the network for different platforms enhanced the interests of safety-critical applications. The reliability evaluation of neural networks are still premature and requires platforms to measure the safety standards required by mission-critical applications. For this reason, the interest in studying the reliability of neural networks is growing. We propose a new approach for evaluating the resiliency of neural networks by using hybrid platforms. The approach relies on the reconfigurable hardware for emulating the target hardware platform and performing the fault injection process. The main advantage of the proposed approach is to involve the on-hardware execution of the neural network in the reliability analysis without any intrusiveness into the network algorithm and addressing specific fault models. The implementation of FireNN, the platform based on the proposed approach, is described in the paper. Experimental analyses are performed using fault injection on AlexNet. The analyses are carried out using the FireNN platform and the results are compared with the outcome of traditional software-level evaluations. Results are discussed considering the insight into the hardware level achieved using FireNN
Cellulose-Based Fibrous Materials From Bacteria to Repair Tympanic Membrane Perforations
No full textPerforation is the most common illness of the tympanic membrane (TM), which is commonly treated with surgical procedures. The success rate of the treatment could be improved by novel bioengineering approaches. In fact, a successful restoration of a damaged TM needs a supporting biomaterial or scaffold able to meet mechano-acoustic properties similar to those of the native TM, along with optimal biocompatibility. Traditionally, a large number of biological-based materials, including paper, silk, Gelfoam®, hyaluronic acid, collagen, and chitosan, have been used for TM repair. A novel biopolymer with promising features for tissue engineering applications is cellulose. It is a highly biocompatible, mechanically and chemically strong polysaccharide, abundant in the environment, with the ability to promote cellular growth and differentiation. Bacterial cellulose (BC), in particular, is produced by microorganisms as a nanofibrous three-dimensional structure of highly pure cellulose, which has thus become a popular graft material for wound healing due to a number of remarkable properties, such as water retention, elasticity, mechanical strength, thermal stability, and transparency. This review paper provides a comprehensive overview of the current experimental studies of BC, focusing on the application of BC patches in the treatment of TM perforations. In addition, computational approaches to model cellulose and TM are summarized, with the aim to synergize the available tools toward the best design and exploitation of BC patches and scaffolds for TM repair and regeneration
On the Analysis of Radiation-induced Failures in the AXI Interconnect Module
No full textDue to the increasing demand for high performance in embedded systems, devices such as SRAM-based programmable devices are becoming an appealing solution to reach high performance with limited costs. However, SRAM-based programmable devices are subjected to various sources of radiation-induced faults that affect their reliability, such as ionizing radiation and particles, even at sea-level. In this paper, we evaluate the reliability of the interconnection module, implemented on the programmable hardware, against radiation-induced faults in the configuration layer. To do so, we performed a fault injection campaign in order to emulate the radiation-induced effects impacting the configuration memory of AP-SoC Zynq 7000, specifically targeting the configuration memory section programming the interconnection module implemented on the programmable logic. This interconnection module is a crucial element for a wide range of applications and mitigation techniques such as hardware-accelerated designs, Dynamic Partial Reconfiguration, or Triple Modular Redundancy; especially if they are adopted to achieve high performance, high bandwidth and high reliability. The fault injection results have been analyzed and classified accordingly with the effect observed on the processor-system side in terms of availability and fault model affecting data computed by cores implemented on the programmable logic side
Poly(lactic acid)-Based Electrospun Fibrous Structures for Biomedical Applications
No full textPoly(lactic acid)(PLA) is an aliphatic polyester that can be derived from natural and renewable resources. Owing to favorable features, such as biocompatibility, biodegradability, good thermal and mechanical performance, and processability, PLA has been considered as one of the most promising biopolymers for biomedical applications. Particularly, electrospun PLA nanofibers with distinguishing characteristics, such as similarity to the extracellular matrix, large specific surface area and high porosity with small pore size and tunable mechanical properties for diverse applications, have recently given rise to advanced spillovers in the medical area. A variety of PLA-based nanofibrous structures have been explored for biomedical purposes, such as wound dressing, drug delivery systems, and tissue engineering scaffolds. This review highlights the recent advances in electrospinning of PLA-based structures for biomedical applications. It also gives a comprehensive discussion about the promising approaches suggested for optimizing the electrospun PLA nanofibrous structures towards the design of specific medical devices with appropriate physical, mechanical and biological functions
Recent advances of polymer-based piezoelectric composites for biomedical applications
No full textOver the past decades, electronics have become central to many aspects of biomedicine and wearable device technologies as a promising personalized healthcare platform. Lead-free piezoelectric materials for converting mechanical into electrical energy through piezoelectric transduction are of significant value in a diverse range of technological applications. Organic piezoelectric biomaterials have attracted widespread attention as the functional materials in the biomedical devices due to their advantages of excellent biocompatibility. They include synthetic and biological polymers. Many biopolymers have been discovered to possess piezoelectricity in an appreciable amount, however their investigation is still preliminary. Due to their piezoelectric properties, better known synthetic fluorinated polymers have been intensively investigated and applied in biomedical applications including controlled drug delivery systems, tissue engineering, microfluidic and artificial muscle actuators, among others. Piezoelectric polymers, especially poly (vinylidene fluoride) (PVDF) and its copolymers are increasingly receiving interest as smart biomaterials due to their ability to convert physiological movements to electrical signals when in a controllable and reproducible manner. Despite possessing the greatest piezoelectric coefficients among all piezoelectric polymers, it is often desirable to increase the electrical outputs. The most promising routes toward significant improvements in the piezoelectric response and energy-harvesting performance of such materials is loading them with various inorganic nanofillers and/or applying some modification during the fabrication process. This paper offers a comprehensive review of the principles, properties, and applications of organic piezoelectric biomaterials (polymers and polymer/ceramic composites) with special attention on PVDF-based polymers and their composites in sensors, drug delivery and tissue engineering. Subsequently focuses on the most common fabrication routes to produce piezoelectric scaffolds, tissue and sensors which is electrospinning process. Promising upcoming strategies and new piezoelectric materials and fabrication techniques for these applications are presented to enable a future integration among these applications
- …
