Driving intention prediction from a bird's-eye view has always been an active research area. However, existing research, on one hand, only focused predicting lane change in highway scenarios and, the other not modeled influence and spatiotemporal relationship of surrounding vehicles. This study extends application to urban road scenarios. A spatial-temporal convolutional long short-term memory (ConvLSTM) model is proposed predict vehicle's lateral longitudinal driving intentions...

Frailty is a common geriatric syndrome associated with many adverse health outcomes. Identifying the risk factors of frailty crucial and insulin resistance (IR) considered as potential target. The estimated glucose disposal rate (eGDR) simple reliable surrogate marker IR. Associations eGDR have not been explored. This study aimed to investigate associations progression. We used data from two prospective cohorts China Health Retirement Longitudinal Study (CHARLS) (HRS). was calculated...

This paper investigates non-orthogonal multiple access (NOMA) system, where the near NOMA user is able to communicate with a base station (BS) directly while far needs resort full-duplex (FD) relay. Wireless power transfer applied our system FD relay integrated splitting architecture can be powered wirelessly by ambient radio signals simultaneously. Moreover, we assume that self-interference cancellation at and inter-user interference are imperfect. For purpose of maximizing energy...

As the wide adoption of in-cloud services (e.g., software-as-a-service), some major identity related issues are brought up. For enterprises, it usually introduces additional cost and risk to manage identities in services. service providers, typical pairwise federation solutions not scalable support single sign-on, composition, etc. among for large environment like cloud. This paper proposes an broker that a trusted third party as trust simplify management user centric manner. With this...

CIT (continuous integration testing) has been widely studied in the testing research field order to start some levels of test as early possible. One challenge lies how simulate behavior those unavailable components. Existing methods like stud and mock fail provide advanced component simulation capabilities required by from perspectives diversified program artifacts, transitivity, configurability. This paper proposes a new apparatus, namely surrogate, address this problem. The surrogate...

Support vector machine (SVM) classifiers are widely applied to hyperspectral image (HSI) classification and provide significant advantages in terms of accuracy, simplicity, robustness. SVM is a well-known learning algorithm that maximizes the minimum margin. However, recent theoretical results pointed out maximizing margin leads lower generalization performance than optimizing distribution, proved distribution more important. In this paper, large (LDM) HSI classification, achieves better...

On consideration of analyzing the relation between drivers' decisions when vehicle-cross process happens in a non-signalized intersection, cooperative driving model is proposed based on reduplicate dynamic game. Considering some characteristics our daily life, profit function established with safety, rapidity and comfort indicators process. Then, multiple Nash Equilibria problem for game analyzed, standard genetic algorithm method adopted to obtain an optimal solution. Finally, this verified...

In contrast to conventional preprocessing aided spatial modulation (PSM), which carries partial information using the indexes of receive antennas, we exploit one antenna implicitly convey and meanwhile harvest energy at remaining antennas. Based on this, propose two n ovel beamforming schemes. The first scheme is maximize sum harvested by receiver. And second minimum receiving power each except for that conveys information. A closed form solution an iterative algorithm are given,...

Geomagnetic positioning technology has proven to be worth investigating in the field of location-based services (LBSs), but geomagnetic alone will generate a certain amount error.To overcome ambiguity single-point data, we developed indoor navigation system, magnetic + particle filters pedestrian dead reckoning (MagPP) based on (PDR) algorithm using smartphone as hardware platform.We calculate measurement trajectory contour PDR solve gross error sequence (MFS).The mean square criterion...

Cracks, potholes, and other defects often occur on infrastructure such as bridges, among which cracks are one of the most frequent defects. They have diverse shapes difficult to detect. Traditional manual inspection methods inefficient low accuracy, while automated machines bulky inconvenient carry use. Based shortcomings existing detection technologies, this paper proposes a portable structural surface crack system based Android platform using handheld image acquisition device. The captures...

Spontaneous renal vessel rupture is a rare clinical emergency. However, pregnancy symptoms and signs are not obvious, the limited examination methods obscure observation. Thus, early challenging to detect, leading misdiagnosis poor prognosis. This paper aims improve clinicians' understanding of this disease reduce rate misdiagnosis.

This paper introduces the basic concepts of improved differential evolution algorithm, industrial control protocol and Fuzzing test technology, as well status quo related research. analyzes structure working characteristics DE protocol. It existing algorithms to make work more efficiently. The is adopted automatically generate valid cases with wider value space. Through two commonly parameters, simulated analyzed, effect algorithm grasped. As a result, tool independent property rights...

Energy operator who can supply electricity, heat and gas to users as an emerging operating entity will play important role in the future energy market. How develop pricing strategy achieving community win-win situation has become issue. In this paper, internet, which includes PV, turbine electric vehicle, aims maximize interests of both users. Based on Stackelberg game, paper proposes pricing, user load response, vehicle charging discharging strategy, genetic algorithm is used obtain...

Removed.

Liver cancer is one of the most common types worldwide. Diagnoses and treatments in early stages liver can help to increase survival rate. However, are based on diagnoses, which require CT scanners as well trained doctors spot tumor from images. The development deep learning provides a possible way fulfill such requirements. In this paper, U-net segmentation algorithm proposed for Various numbers filters were set check their improvements model prediction accuracies along with three different...

Most of the indoor positioning algorithms based on magnetic data mainly focus reducing accumulated error odometry data, such as signals produced by inertial sensors. However, in most cases using smartphones environment, those approaches seem unfeasible due to absence Thus, this paper, we try study a algorithm exclusively data. We refer some thinking from steps Particle Filter and conduct an experiment verify application new algorithm. Besides, use variance result previous step decrease area...

In order to improve the application security of smart card, this paper has proposed an encryption scheme card combined with AES algorithm and high-speed key generation algorithm. Furthermore, characteristics M1 introduced work program implemented program.