Abnormal falls in public places have significant safety hazards and can easily lead to serious consequences, such as trampling by people. Vision-driven fall event detection has the huge advantage of being non-invasive. However, actual scenes, behavior is rich diversity, resulting strong instability detection. Based on study stability human body dynamics, article proposes a new model posture representation behavior, called "five-point inverted pendulum model", uses an improved two-branch...

Today, with the rapid development of artificial intelligence, dubbing industry has ushered in brand-new opportunities and challenges. However, traditional teaching mode major significant deficiencies cultivation personalized expression emotional resonance ability, which makes it difficult for students to meet current employment environment market needs. This paper aims propose a series innovative countermeasures by analyzing shortcomings mode. These can not only improve effectiveness...

The developments in wireless sensor network (WSN) that enriches with the unique capabilities of cognitive radio technique are giving impetus to evolution Cognitive Wireless Sensor Network (CWSN). In a CWSN, nodes can opportunistically transmit on vacant licensed frequencies and operate under strict interference avoidance policy other users. However, typical constraints energy conservation from battery-driven design, local spectrum availability, reachability nodes, large-scale architecture...

Oil chromatography data together with its variation trend provide the key basis for evaluation of transformer health state. The existing studies on deep belief network (DBN) and support vector machine (SVM) have reported a few results in field oil prediction. However, above-mentioned methods are proposed classification problem, so there is no theoretical applying to time-series wrong usage limits accuracy predicted results, which observed as obvious "time-shift" error prediction curve,...

For large-scale iris recognition tasks, the determination of classification thresholds remains a challenging task, especially in practical applications where sample space is growing rapidly. Due to complexity samples, threshold difficult determine with increase samples. The key issue solving such problems obtain feature vectors more obvious discrimination. Therefore, we train deep convolutional neural networks based on large number samples extract features. More importantly, an optimized...

In order to realize the reliable and safe operation of smart railways, provide high quality information transmission service for passengers, railway system needs develop innovative communication network advanced technology meet gradually increasing demand multi-dimensional comprehensive resources. The Space-Air-Ground Integrated Network ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SAGIN</i> ) can seamless services land, sea, air space...

Noise robustness is essential for deploying automatic speech recognition (ASR) systems in real-world environments. One way to reduce the effect of noise interference employ a preprocessing module that conducts enhancement, and then feed enhanced an ASR backend. In this work, instead suppressing background with conventional cascaded pipeline, we noise-robust representation learned by refined self-supervised framework noisy recognition. We propose combine reconstruction contrastive learning...

With the continuous development of intelligent transportation industry, target tracking has become an important research direction. Under normal circumstances, due to complex road environment and changing backgrounds, millimeter wave radar more interference when detecting targets. In addition variety targets in different scattering intensity multiple parts, flicker noise on must be considered. The combination these noises can affect accuracy measurement even make lose for a short time. paper...

The high-speed railway wireless communication network architecture is gradually transforming from a traditional GSM-Railway (GSM-R) cellular to 5G heterogeneous ultra-dense network. How effectively deal with the interference while obtaining capacity gain has become an inevitable problem. This article proposes power allocation alignment algorithm based on game equilibrium goal of joint optimization about throughput and energy efficiency for imperfect channels in communication. Firstly,...

Voice-controlled house-hold devices, like Amazon Echo or Google Home, face the problem of performing speech recognition device-directed in presence interfering background speech, i.e., noise and from another person media device proximity need to be ignored. We propose two end-to-end models tackle this with information extracted anchored segment. The segment refers wake-up word part an audio stream, which contains valuable speaker that can used suppress noise. first method is called...

Wireless communication plays a vital role in the operations of modern rail transportation. The rapid motion characteristics train make wireless spectrum environment unstable and discontinuous. These uncertainties, coupled with inherent scarcity spectrum, lead to inefficiencies railroad communications. application cognitive radio is becoming cutting-edge research field railway communication. This paper first analyzes physical infrastructure network determines base station as key node...

In the evolution of railway mobile communications from Long Term Evolution for Railway ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">LTE-R</i> ) to future 5th Generation Wireless System xmlns:xlink="http://www.w3.org/1999/xlink">5G</i> ), rapid increase in number low-power base station nodes along has brought more frequent handovers. The current handover parameter selection mechanism often relies on on-site measured results a limited...

High-speed rails have a significantly higher driving safety requirement than other public transport because of more passengers and faster speed. However, the particularity train often leads to railway drivers susceptible drowsiness fatigue, this fatigue has distinct personality. Here, authors analyse mechanism individual generation different drivers, use eye movement data collected by non-contact means as an objective measurement, combine subjective sleepiness assessment reveal existence...

Cemented paste backfill (CPB) is an economic and environmental friendly technique applied in underground mining for supporting surrounding rock replacing the pillar. However, little known about temperature effects on CPB mines having a large fluctuation. The main purpose of this research was to investigate effect change rheological properties (e.g., shear stress apparent viscosity) with copper mine tailings. Specifically, series tests were conducted 6 samples using Brookfield R/S+ Rheometer...

A radical salt experiences crystal–crystal and crystal–plastic crystal phase transitions with magnetic bistability negative thermal expansion, high room-temperature ion-conduction in superionic conduction plastic phase.

Medium access control (MAC) protocols have a large impact on the achievable system performance for wireless ad hoc networks. Because of limitations existing analytical models networks, many researchers opted to study MAC via discrete-event simulations. However, as network scenarios, traffic patterns and physical layer techniques may change significantly, simulation alone is not efficient get insights into impacts performance. In this paper, we analyze IEEE 802.11 distributed coordination...