Edge accelerator is a class of brand-new purpose-built System On Chip (SoC) for running deep learning models efficiently on edge devices. These accelerators offer various benefits such as ultra-low latency, sensitive data protection, and high availability due to their locality are opening up interminable opportunities building sensory systems in the real world. Naturally, context awareness systems, e.g., IoT, wearables, other devices, emergence pushing us rethink how we design these at...

Direction finding and positioning systems based on RF signals are significantly impacted by multipath propagation, particularly in indoor environments. Existing algorithms (e.g MUSIC) perform poorly resolving Angle of Arrival (AoA) the presence or when operating a weak signal regime. We note that digitally sampled frontends allow for easy analysis signals, their delayed components. Low-cost Software-Defined Radio (SDR) modules enable Channel State Information (CSI) extraction across wide...

The paper explores the possibility of using wrist-worn devices (specifically, a smartwatch) to accurately track hand movement and gestures for new class immersive, interactive gesture-driven applications. These applications need two special features: (a) ability identify from continuous stream sensor data early--i.e., even before gesture is complete, (b) precisely hand's trajectory, though underlying inertial noisy. We develop approach that tackles these requirements by first building...

Energy harvesting, from a diverse set of modes such as light or motion, has been viewed the key to developing batteryless sensing devices. In this paper, we develop nascent idea harvesting RF energy WiFi transmissions, applying it power prototype wearable device that captures and transmits accelerometer sensor data. Our solution, WiWear, two innovations: 1) beamforming transmissions significantly boost receiver can harvest ~23 meters away, 2) smart zero-energy, triggering inertial sensing,...

Thermal imagery is emerging as a viable candidate for 24-7, all-weather pedestrian detection owning to thermal sensors’ robust performance under different weather and illumination conditions. Despite the promising results obtained from combining visible (RGB) cameras in multi-spectral fusion techniques, complex synchronization requirements, including alignment calibration of sensors, impede their deployment real-world scenarios. In this paper, we introduce novel approach domain adaptation...

Recent Network Intrusion Detection System (NIDS) utilizes more and Regular Expression to describe malicious patterns existing in the content payload of packets. Many researches are investigated several techniques introduced optimize performance support all functions regular expression on hardware platform. However there is very few minimization multiple expressions. This paper takes account compiling expressions with respect resources. We take advantage block memory implement character...

Rapid development of network makes it a very important and vulnerable part every field life. Many intrusion detection systems are developed to protect the using signature-based matching technique. For connection oriented protocols, such as Transmission Control Protocol, data should be reassembled before being scanned by engine. Several techniques introduced reassemble TCP packets on FPGA. However, they have some disadvantages inefficient memory, unscalable system, unsupported complex...

The synthesis and cytotoxicity evaluation of fourteen new chalcones containing 5-fluorouracil were introduced in which nine as Mannich bases 8a-i synthesized from 2'-hydroxyacetophenone (1) three steps including chloromethylation, N-alkylation Claisen-Schmidt reactions. Other 14a-e obtained Click reaction between 1,3-dipropargyl-5-fluorouracil azide derivatives 2',4'-dihydroxychalcone. bio-assay results showed that 11 (excluding 8d, 8f 8h) exhibited against four human cancer cell lines...

Ubiquitous positioning for pedestrians in adverse environments has been a long standing challenge. Despite dramatic progress made by Deep Learning, multi-sensor deep odometry systems still pose high computational cost and suffer from cumulative drifting errors over time. Thanks to the increasing power of edge devices, we propose novel ubiquitous solution integrating state-of-the-art models on with an EKF (Extended Kalman Filter)-LoRa backend. We carefully select compare three sensor...

Ultra-Wide-Band (UWB) ranging sensors have been widely adopted for robotic navigation thanks to their extremely high bandwidth and hence resolution. However, off-the-shelf devices may output ranges with significant errors in cluttered, severe non-line-of-sight (NLOS) environments. Recently, neural networks actively studied improve the accuracy of UWB using channel-impulse-response (CIR) as input. previous works not systematically evaluated efficacy various packet types possible combinations...

The paper introduces a novel, holistic approach for robust Screen-Camera Communication (SCC), where video content on screen is visually encoded in human-imperceptible fashion and decoded by camera capturing images of such content. We first show that state-of-the-art SCC techniques have two key limitations in-the-wild deployment: (a) the decoding accuracy drops rapidly under even modest extraction errors from captured images, (b) they generate perceptible flickers common refresh rate screens...

Twenty-one new chalcones 9a-m (excluding 9e, 9j and 9l) 10a-m 10j 10l), containing nucleobases were synthesized from 2'-hydroxyacetophenone (1) by the reactions including chloromethylation, nucleophilic substitution with thymine uracil, Claisen-Schmidt reactions. These evaluated for in vitro cytotoxicity against five human cancer cell lines: SK-LU-1, Hep-G2, MCF7, SW480 P388. The results showed that most of tested exhibited inhibitory activity lines except 10h, 10i. Among chalcones, compound...

Nowadays, networking plays a very important role in businesses. The network infrastructure has been developing rapidly. Moreover, the invention of TCP protocol motivates lot applications. Many these applications require line rate processing packets and reassembly for stateful processing. Several techniques are introduced to reassemble on FPGA. However, they have some disadvantages such as inefficient memory, unscalable system, unsupported complex connections. In this paper, we propose...

We describe our vision of a multiple mobile or wearable device environment and share initial exploration in multi-wrist gesture recognition. explore how multi-device input output might look, giving four scenarios everyday use that show the technical challenges need to be addressed. system which allows for recognition distributed between devices, fusing streams on resource-rich (e.g., phone). An Interactor layer recognises common gestures from fusion engine, provides abstract scrolling...

Non-enzymatic glucose sensors have gained significant attention owing to their potential for accurate and cost-effective detection. Nanomaterials play a pivotal role in enhancing performance of non-enzymatic sensor. In this study, novel sensor based on Ni(OH)2/NF (nickel hydroxide/nickel foam) materials was developed using different pulse voltammetry technique. Chemical precipitation methods been used grow Ni(OH)2 nanostructures directly nickel foam electrode. The characterization the...

The paper presents a novel synthesis of AZT modified derivatives via click chemistry. Firstly, the reaction 2,4-dihydroxyacetophenone (1) with propargyl bromide in acetone gave intermediates 2 75% yield. Next, new chalcones 3a-g were obtained 55-65% yields through Claisen-Schmidt different aromatic aldehydes. These then oxidatively cyclized DMSO presence I2 to furnish flavones 4a-g 45-59% yields. Finally, title compounds 5a-g and 6a-g 3a-h 4a-h 3’-Azido-2’- deoxythymidine (AZT). Novel 5a-g,...

An efficient synthesis of novel AZT derivatives possessing Mannich base moieties via click chemistry is described. Firstly, the reaction ferulic acid (1) with propargyl bromide and amine gave intermediates 2 3, respectively. Secondly, 3 secondary amines in ethanol under reflux conditions furnished two series bases 2a-h 3a-h good yields. Finally, title compounds 5a-h 6a-h were obtained 3’-Azido-2’-deoxythymidine (AZT) moderate to Novel evaluated for vitro cytotoxic activity against three cell...

In this paper, we propose an automated deep learning-based software, especially for reconstructing 3D lung images and estimating Total Lung Volume (TLV). The system is mainly designed Vietnamese users with as the default language. purpose of our study to build based on learning machine-learning models measure TLV construct a prototype lung. training data collected from Cancer Imaging Archive (TCIA) dataset, provided by 2017 CT Segmentation Challenge testing proposed model. Our utilizes...

Ultra-low power communication is critical for supporting the next generation of battery-operated or energy harvesting battery-less Internet Things (IoT) devices. Duty cycling protocols and wake-up receiver (WuRx) technologies, their combinations, have been investigated as energy-efficient mechanisms to support selective, event-driven activation In this paper, we go one step further show how WuRx can be used an efficient multi-purpose low downlink (LPD) channel. We demonstrate (a) extend...

Direction finding and positioning systems based on RF signals are significantly impacted by multipath propagation, particularly in indoor environments. Existing algorithms (e.g MUSIC) perform poorly resolving Angle of Arrival (AoA) the presence or when operating a weak signal regime. We note that digitally sampled frontends allow for easy analysis signals, their delayed components. Low-cost Software-Defined Radio (SDR) modules enable Channel State Information (CSI) extraction across wide...

We tackle the problem of developing a wearable device that operates indoors on harvested RF energy, but can support gesture tracking applications require relatively energy-intensive inertial sensors. propose an infrastructure-assisted energy harvesting paradigm, where ubiquitously-deployed WiFi infrastructure helps to significantly improve power, by employing state-of-the-art techniques for device-free user localization and beamformed, directional packet transmission. Preliminary...

First-person captioning is significant because it provides veracious descriptions of egocentric scenes in a unique perspective. Also, there need to caption the scene, a.k.a. life-logging, for patients, travellers, and emergency responders an narrative. Ego-captioning indeed non-trivial since (1) Ego-images can be noisy due motion angles; (2) Describing scene first-person narrative involves drastically different semantics; (3) Empirical implications have made on top visual appearance...

No abstract available.