In this letter, we study the on-demand UAV-BS placement problem for arbitrarily distributed users. This is modeled as a knapsack-like problem, which NP-complete. We propose density-aware algorithm to maximize number of covered users subject constraint minimum required data rates per user. Simulations are conducted evaluate performance proposed in real environment with different user densities. Our numerical results indicate that various densities our solution can service more guaranteed...

We present an enhanced segmentation method to reduce the interference of shadows for vehicle detection. The main advantage proposed is its low-complexity. use only luminance image shadow removal and keep chrominance components intact. current each pixel compared with a pixel-dependent threshold locating regions. With that, regions can be located more accurately moving objects extracted completely. Experimental results verify approach show that it helpful

Deep learning involves a difficult nonconvex optimization problem with large number of weights between any two adjacent layers deep structure. To handle data sets or complicated networks, distributed training is needed, but the calculation function, gradient, and Hessian expensive. In particular, communication synchronization cost may become bottleneck. this letter, we focus on situations where model distributedly stored propose novel Newton method for neural networks. By variable...

An edge data center (EDC) network infrastructure is required to flexibly deliver massive bandwidth and ultralow latency for a 5G cloud. Existing electrical-switching-based infrastructure, however, has been shown incapable of meeting such requirements. In this paper, we present an EDC architecture prototype testbed, referred as the intelligence-defined optical tunnel system (OPTUNS). OPTUNS consists set switching subsystems that operate collectively facilitate packet transport through logical...

Veterinary staff could be in frequent exposure to a variety of physical, chemical and biological hazards. Epidemiologic studies have showed that veterinarians are at increased risk many occupational illnesses such as dermatoses, respiratory tract illnesses, zoonotic infections, acute pesticide-associated toxicity, fetal loss site-specific cancers. Most hazards distributed the form aerosols workplaces. This study intended investigate potential faced by small-animal practices through field...

Abstract This study examined the hydrodynamic characteristics of a liquid–solid fluidized‐bed bioreactor using elastic particles (PVA gel beads) various diameters as carriers. The drag coefficient‐Reynolds number, velocity‐voidage, and expansion index‐Reynolds number relationships observed during fluidization PVA beads in fluidized bed our experiments were compared with published results. Predictions made from previous correlations new experimental findings, revealing inadequacy most these...

Filter bank multi-carrier based on offset quadrature amplitude modulation (FBMC/OQAM) suffers from inter-symbol interference and inter-subchannel in multi-path fading channels. By exploiting the special structures of FBMC/OQAM signal resulting correlation matrix assuming perfect estimation channel noise power, we study properties decision feedback equalizer (DFE) designed by minimum mean square error criterion. The leads to efficient mathematically equivalent computation DFE coefficients....

Taiwanese green propolis is rich in prenylated flavonoids and exhibits a broad range of biological activities, such as antioxidant, antibacterial, anticancer ones. The bioactive compounds are propolins, namely C, D, F, G. concentration propolins varies depending on the season geographic location. Thus, it critical to establish standard repeatable procedure for determining quality propolis. Here, we present protocol that uses ethanol-based extraction, high-performance liquid chromatography,...

In this paper, an NC processor for power-efficient real-time universal visual recognition is proposed with following features: 1) A grey matter-like homogeneous many-core architecture event-driven hybrid MIMD execution provides 1.0TOPS/W efficient acceleration operations; 2) white Kautz NoC 2.3Tb/s throughput, fault/congestion avoidance and redundancy-free multicast 151Tb/s/W power efficiency, which 2.7-3.9x higher than previous NoC-based processors.

Mechanisms underlying the pathogenesis of psychotic disorders were explored by monitoring expression GABAergic neurons in an animal model. Male rats postnatal days 21 and 60 intraperitoneally injected with amphetamine (Amph), 5 mgkg, or saline three times daily for 6 d. After 1-d 14-d withdrawal from Amph, they challenged on day 8 (W1d) (W14d) a single same dosage then perfused. Immunostaining brain sections using anti-glutamic acid decarboxylase (GAD67) antiserum revealed that Amph...

We present the architecture/testbed of an SDN-based optical-tunnel-network system (OPTUNS) for 5G edge-datacenters. Benchmarking results show that, compared to electrical-spine-leaf datacenter-network, OPTUNS achieves 82.6% power saving and significantly lower mean/p99 latency under high traffic load/locality.

This letter reports the fabrication of polycrystalline silicon (poly-Si)tunneling field effect transistors(Tunneling FETs) using green nanosecond laser crystallization (GLC). During GLC process, Si is full-melted by scanning, hence recrystallized poly-Si thin films with grain size as large 1.2 μm are attained. makes it possible to fabricate tunneling FET high-quality films. Compare FETs fabricated solid phase (SPC), ones via show better subthreshold swing (S.S) 418 mV/dec. and larger on/off...

Taiwanese green propolis is rich in prenylated flavonoids and exhibits a broad range of biological activities, such as antioxidant, antibacterial, anticancer ones. The bioactive compounds are propolins, namely C, D, F, G. concentration propolins varies depending on the season geographic location. Thus, it critical to establish standard repeatable procedure for determining quality propolis. Here, we present protocol that uses ethanol-based extraction, high-performance liquid chromatography,...

This paper was aimed to investigate the influence of online coding courses on logical reasoning ability towards fifth and sixth graders. Students used "Code.org programming courses" about three months. Through "Raven's Standard Progressive Matrices" before after testing understand change students' in experiment. And based "Technology Acceptance Model"' researcher edit Learning Attitude Questionnaire" explore attitude students factors influencing learning courses.

Sleep endoscopy is a valuable diagnostic tool for assessing airway obstruction during sleep. Precise segmentation of structures essential clinical evaluation and treatment planning. In this study, we utilize well-known deep learning method, U-Net++, automatic velopharyngeal in sleep images. Here U-Net++ equipped with an attention mechanism to enhance the representational power model, focus loss handle imbalanced situation areas, data augmentation techniques introduce heterogeneity then get...

In this investigation, an adaptive reset strategy of a 10-bit SAR-ADC particularly designed for the applications in FOG (fiber optic gyroscope) systems is demonstrated with TSMC 40 nm CMOS technology. The proposed approach takes advantage high correlation phase shift generated by continuous signals such that MSB bits two consecutive samples likely remain same, which implies there no need to re-converted. By replacing conventional all-reset method adaptive-reset strategy, throughput SAR ADC...

Deep learning involves a difficult non-convex optimization problem with large number of weights between any two adjacent layers deep structure. To handle data sets or complicated networks, distributed training is needed, but the calculation function, gradient, and Hessian expensive. In particular, communication synchronization cost may become bottleneck. this paper, we focus on situations where model distributedly stored, propose novel Newton method for neural networks. By variable...

In this paper, we propose two new interdigital electrode designs, flat and circular, to increase the signal level detection sensitivity of laser-induced plasma.