i-ETC : ISEL Academic Journal of Electronics Telecommunications and Computers
Not a member yet
54 research outputs found
Sort by
Reducing Execution Time in FaaS Cloud Platforms
Full text linkThis article research into the significance of caching within Function-as-a-Service (FaaS) environments, exploring how caching strategies can substantially enhance performance and scalability in the realm of serverless computing. A versatile caching architecture for FaaS is introduced, tailored to accommodate different caching strategies. The architecture is implemented by extending an open-source FaaS framework, specifically Google's Functions Framework. An aspect-oriented approach is adopted to transparently specify the relevant objects that should be cached, effectively decoupling function implementation from deployment configuration. The study extensively investigates various caching mechanisms, encompassing in-process, out-of-process, and network caching, and systematically assesses their impact on response times and resource utilization. The findings underscore the trade-offs inherent in employing caching techniques, ultimately aiming to optimize FaaS performance and improve overall system efficiency
Lightweight and Efficient Architecture for AES Algorithm based on FPGA
No full textDifferent platforms, such as resource limited devices and high-performance processors, are used in IoT networks, each with its own set of resource, performance, and security needs. It is critical to optimize existing standard cryptographic algorithms to meet the needs of today's networks, yet this is a difficult undertaking. In this paper, a compact and efficient architecture for the Advanced Encryption Standard (AES) is developed and implemented using several FPGA platforms, with the goal of addressing both restricted and high-performance platforms in IoT networks. To create compact and efficient AES based on FPGA, a hybrid optimization technique is applied. The implementation makes advantage of FPGA embedded resources such as BRAMs and DSP slices. To synthesize and implement it on the Xilinx Virtex-7 device, the Vivado HLS tool 2019.1 is utilized. Similarly, as devices older than the Xilinx 7-series platforms are not directly supported by Vivado HLS tool, Xilinx 14.5 ISE tool is used to synthesize and implement it. Smaller resources, such as 572 slices, 8 BRAMs, and 32 DSP slices, are used in comparison to the implementation outcomes found in literature. Additionally, improved throughput performance (112.399 Gbps) was achieved by satisfying the current work's optimization targets
An Approach to Estimate Electric Vehicle Driving Range
No full textThe use of electric vehicle (EV) has grown rapidly over the past few years. The EV is now accepted as a reliable and eco-friendly means of transportation. When choosing an EV, usually one of the key parameters of choice for the customer is its driving range (DR) capability. This is a decisive factor since it minimizes the drivers anxiety on a trip. The DR depends on many factors that must be taken into account when attempting its prediction.In this paper, we explore the use of machine learning (ML) techniques to estimate the DR prediction.We use regression techniques on models trained with publicly available datasets, evaluated with standard metrics.The prediction results are better than those provided by statistical techniques, thus being quite encouraging.As the end result, we also provide a ML benchmark written in Python, aiming to advance future research on this topic
A Fatigue and Drowsiness Detection System Using Inertial Sensors and Electrocardiogram Signals
No full textThe interest in monitoring a drivers conditions and performance has increased in the past years, to make the roads safer both for drivers and pedestrians. This raised the idea of developing a system to monitor the drivers conditions to prevent road disasters. In this paper, we propose a system to monitor the drivers fatigue and drowsiness, based on the Car- dioWheel system, developed by CardioID. The proposed system records both the persons ECG signal and the motion of the steering wheel during the driving session. The amount of data acquired demands a compression stage for transmission with the goal to reduce the required bandwidth. The transmission of the compressed data is done via Bluetooth Low Energy, with an exclusive profile developed for this system. To detect fatigue and drowsiness patterns, a machine learning approach was taken. Among the evaluated classifiers, the Support Vector Machines technique proved to be the best classification method with the highest accuracy. Thus, the developed prototype has the ability to warn the driver about his physiological and physical states, increasing the safety in the roads
Simulation of plasmonic effects of Metal (Au,Ag and Al) NPs and rGO embedded in aqueous solutions
Full text linkGraphene [1] is a material that has been extensively explored in recent years as a material with optical properties that enable its application as active material in sensing devices.In this work we will study plasmonic effects and optical properties of graphene and metal nanoparticles (AuNPs), comparing its results, whenever possible, with results obtained in previous studies. Analysis will be supported by simulation results obtained with Matlab (“Mie analysis”)
Combined detection of nitrite and bioelectrical activity using microelectrode arrays and a phosphate buffered saline solution
Full text linkThis paper presents the manufacturing of amicroelectrode array, on printed circuit boards (PCB) and onsilicon substrate, composed of 60 to 64 gold microelectrodes(between 4 µm and 70 µm each in diameter). The sensor has ameasurement total area of 0.5 cm in radius, one referenceelectrode (1 mm2 in area) and 28.4 mm wide and 28.4 mm long.These microelectrodes were used for checking and logging ofextracelullar local field potential of cell culture and nitritemeasurement in a phosphate buffered saline solution(electrolytical aqueous medium). In addition, an apparatus toshield from electromagnetic interference for connecting thearrays was designed to allow the capture of electrochemicalreactions or electronic signals by the microelectrodes, forexample: nitrite or cardiac potential measurement, respectively.Finally, biocompatibility tests of the array structures wereperformed. The preliminary electrical and biocompatibilitytesting, along with the collected data, has shown promisingresults pointing to the development of an accurate sensor afterthe completion of this study. The sensor has potentially abroader range of applications with only a few adaptations anddue its good accuracy it can be a very useful resource for manychemical and biological applications
Urban Sound Event Detection and Classification
No full textThe ability to automatically detect/classify complex and dynamic urban sounds is important tool for urban planning such as building efficient noise monitoring and control, surveillance systems, urban soundscape mappings, and could be a foundation for improvement of life in the cities of tomorrow.We present a proof of concept for a sound detection, monitoring and classification system for urban environments in the context of Smart Cities. Ultimately, the proposed system be deployed in numerous urban locations for long periods of time, allowing for the collection of urban acoustic data
A Deep Learning Approach to Identify Not Suitable for Work Images
Full text linkWeb Archiving (WA) deals with the preservation of portions of the World Wide Web (WWW) allowing their availability for future access. Arquivo.pt is a WA initiativeholding a huge amount of content, including image files.However, some of these images contain nudity and pornography, that can be offensive for the users, and thus being Not SuitableFor Work (NSFW). This work proposes a methodology to classify NSFW images available at Arquivo.pt, using deep neural network approaches. A large dataset of images is built using Arquivo.pt data and two pre-trained neural network models, namely ResNet and SqueezeNet, are evaluated and improved for the NSFW classification task, using the dataset.The evaluation of these models reported an accuracy of 93% and 72%, respectively. After a fine tuning stage, the accuracy of these models improved to 94% and 89%, respectively.The proposed solution is integrated into the Arquivo.pt Image Search System, enabling the filtering of the problematic NSFW images. At the time of this writing, the proposed solution is in production at https://arquivo.pt/images.js
APC speech coding techniques applied to ECG signals
No full textThis paper describes an ECG signal coding with an adaptive predictive coding scheme using speech techniques like linear predictive coding and long-term prediction
Applications for a-Si:H TFTs: Modelling and Simulation
Full text linkHydrogenated amorphous silicon thin film transistors have been used as switching elements in liquid crystal displays and large area matrix addressed sensor arrays. Later, these devices have also been used as analogue active elements in organic light emitting diode displays. However, this technology suffers from bias induced meta-stability. This issue introduces both threshold voltage and subthreshold slope shifts over time when gate bias is applied. Such instabilities jeopardize long term performance of circuits that rely on these components. Nevertheless, hydrogenated amorphous silicon thin film transistors present an exponential transfer characteristic when operating on subthreshold region and their typical power consumption is under 1 µW. This low power characteristic makes these devices ideally suited for low power electronic design.This work demonstrates, through transient analysis of a well-established simulation model for hydrogenated amorphous silicon, the viability of thin film transistors technology to perform both analogue and digital functions. Hence, these structures may be used in both application fields. To this end, two different sets of analyses have been conducted with hydrogenated amorphous silicon based thin film transistors. The first set considers a driving circuit for an active matrix of organic light emitting diodes, biased in a way to minimize the “memory effect” (increasing shift on threshold voltage) due to long term operation. The second set of analyses were conducted upon the implementation of complementary output universal gates, namely NOR/OR and XNOR/XOR elements