Abstract. Advances in embedded systems and low-cost gas sensors are enabling a new wave of air quality monitoring tools. Our team has been engaged the development low-cost, wearable, monitors (M-Pods) using Arduino platform. These M-Pods house two types – commercially available metal oxide semiconductor (MOx) used to measure CO, O3, NO2, total VOCs, NDIR CO2. The MOx low cost show high sensitivity near ambient levels; however they display non-linear output signals have cross-sensitivity...

Passive indoor localization is important. Unlike active techniques, it does not require for users to carry measuring devices, e.g., smart phones. Thus, widely used in applications such as security, housing, object tracking, etc. However, real-world applications, the passive accuracy limited due environment noises, multipath effect, To address those problems, this paper, we propose use channel state information (CSI) instead. Specifically, make following contributions: 1) design a CSI-based...

People spend approximately 70% of their time indoors. Understanding the indoor environments is therefore important for a wide range emerging mobile personal and social applications. Knowledge floorplans often required by these However, are either unavailable or obtaining them requires slow, tedious, error-prone manual labor.

Deep neural networks are becoming popular and important assets of many AI companies. However, recent studies indicate that they also vulnerable to adversarial attacks. Adversarial attacks can be either white-box or black-box. The assume full knowledge the models while black-box ones none. In general, revealing more internal information enable much powerful efficient in most real-world applications, embedded devices is unavailable. Therefore, this brief, we propose a side-channel based...

Cherry tomato (Solanum lycopersicum) is popular with consumers over the world due to its special flavor. Soluble solids content (SSC) and firmness are two key metrics for evaluating product qualities. In this work, we develop non-destructive testing techniques SSC fruit based on hyperspectral images corresponding deep learning regression model. Hyperspectral reflectance of 200 fruits derived spectrum ranging from 400 1,000 nm. The acquired corrected spectral information extracted. A novel...

Abstract The freshness of vegetable soybean (VS) is an important indicator for quality evaluation. Currently, deep learning-based image recognition technology provides a fast, efficient, and low-cost method analyzing the food. RGB (red, green, blue) widely used in study food appearance In addition, hyperspectral has outstanding performance predicting nutrient content samples. However, there are few reports on research classification models based fusion data these two sources images. We...

Most people spend more than 90% of their time indoors; indoor air quality (IAQ) influences human health, safety, productivity, and comfort. This paper describes MAQS, a personalized mobile sensing system for IAQ monitoring. In contrast with existing stationary or outdoor systems, MAQS users carry portable, location tracking sensors that provide information. To improve accuracy energy efficiency, incorporates three novel techniques: (1) an accurate temporal n-gram augmented Bayesian room...

The reliability of multi-processor systems-on-chip (MPSoCs) is affected by several inter-dependent system-level and physical effects. Accurate fast modeling a primary challenge in the design optimization reliable MPSoCs. This paper presents framework that integrates device-, component-, models. contains modules for electromigration, time-dependent dielectric breakdown, stress migration, variable-amplitude thermal cycling. A new statistical distribution proposed accurate characterization...

Flexible humidity sensors with high sensitivity, fast response time, and outstanding reliability have the potential to revolutionize electronic skin, healthcare, non-contact sensing. In this study, we employed a straightforward nanocluster deposition technique fabricate resistive sensor on flexible substrate, using molybdenum oxide nanoparticles (MoOx NPs). We systematically evaluated humidity-sensing behaviors of MoOx NP film-based found that it exhibited exceptional sensing capabilities....

System reliability is a crucial concern especially in multicore systems which tend to have high power density and hence temperature. Existing reliability-aware methods are either slow non-adaptive (offline techniques) or do not use task assignment scheduling compensate for uneven core wear states (online techniques). In this article, we present dynamically-activated algorithm based on theoretical results that explicitly optimizes system lifetime. We also propose data distillation method...

Indoor air quality is important. It influences human productivity and health. Personal pollution exposure can be measured using stationary or mobile sensor networks, but each of these approaches has drawbacks. Stationary network accuracy suffers because it difficult to place a in every location people might visit. In drift resistance are generally sacrificed for the sake mobility economy. We propose hybrid architecture, which contains both sensors (for accurate readings calibration)...

Magnetic induction tomography (MIT) is an efficient solution for long-term brain disease monitoring. It focuses on reconstructing the brain's bio-impedance distribution through non-intrusive electromagnetic fields. However, high-quality reconstruction of images remains a significant challenge, as from weak and noisy signals highly non-linear ill-conditioned problem. In this work, we propose generative adversarial network (GAN) enhanced MIT technique, named MITNet, based complex convolutional...

Mobile sensing systems carried by individuals or machines make it possible to measure position- and time-dependent environmental conditions, such as air quality radiation. The low-cost, miniature sensors commonly used in these are prone measurement drift, requiring occasional re-calibration provide accurate data. Requiring end users periodically do manual calibration work would many mobile impractical. We therefore argue for the use of collaborative, automatic among nearby sensors, solutions...

This paper investigates and analyzes three typical elevation beamforming scenarios which are most likely to be applied in future LTE-Advanced systems: vertical sectorization with same carrier frequency, different frequency based on aggregation, user-specific beamforming. Preliminary evaluation using WINNERII/WINNER+ 3D MIMO channel modeling is carried out compare these the conventional system single downtilting. It shown that latter two can achieve good performance.

Abstract. Advances in embedded systems and low-cost gas sensors are enabling a new wave of low cost air quality monitoring tools. Our team has been engaged the development wearable monitors (M-Pods) using Arduino platform. The M-Pods use commercially available metal oxide semiconductor (MOx) to measure CO, O3, NO2, total VOCs, NDIR CO2. MOx show high sensitivity near ambient levels; however they display non-linear output signals have cross effects. Thus, quantification system was developed...

Passive indoor localization techniques can have many important applications. They are nonintrusive and do not require users carrying measuring devices. Therefore, widely used in critical areas, such as security, logistics, healthcare, etc. However, because of the unpredictable environment dynamics, existing be quite inaccurate, which greatly limits their real-world To address those problems, this work, we develop a channel state information (CSI) based technique. Unlike methods, employ both...

Animals search for food based on certain optimal principles and over time form foraging patterns effective survival in changing environments. Due to the many choices available modern society, we also face a decision where get their food. We call this "modern human foraging," since Internet makes much more convenient than before. People online venues, or restaurants, through websites such as Yelp, write reviews they tasted, which turn, facilitate others' searches future. These activities make...

Deep neural networks are becoming increasingly popular. However, they also vulnerable to adversarial attacks. The existing attack methods include white-box and black-box attack. assumes full model knowledge while the one none. In this brief, we propose a novel method between these two. Specifically, have made following contributions: (1) gray-box attack, which utilizes side-channel predict structure based on pre-trained classifier (2) validate our real-world experiments. experimental results...

Remote monitoring service for elderly persons is important as the aged populations in most developed countries continue growing. To monitor safety and health of population, we propose a novel omni-directional vision sensor based system, which can detect track object motion, recognize human posture, analyze behavior automatically. In this work, have made following contributions: (1) develop remote system provide real-time automatic care (2) design motion history or energy images algorithm...

One major challenge for modern artificial neural networks (ANNs) is that they typically does not handle incremental learning well. In other words, while the new features, performances of existing features usually deteriorate. This phenomenon called catastrophic forgetting, which causes great problems continuous, incremental, and intelligent learning. this work, we propose a dynamic correction vector based algorithm to address both bias problem from knowledge distillation overfitting problem....