Resent years, with the rapid development of embedded device industry, embedded mobile devices have been popular. The power of these devices is supplied by batteries, but battery capacity is limited by its volume and w...
详细信息
ISBN:
(纸本)9781467308946;9788994364261
Resent years, with the rapid development of embedded device industry, embedded mobile devices have been popular. The power of these devices is supplied by batteries, but battery capacity is limited by its volume and weight. So the power problem of embedded mobile devices has been a research focus. This paper analyzes the power consumption when embedded devices make handoff (switch from one network to another) in WiFi networks and propose the Power Preference Handoff technology (PPHT) as an optimization scheme. According to the experiment results, we demonstrate that embedded mobile devices making frequent handoff in WiFi networks will obviously increase the power consumption, and the PPHT scheme can reduce substantial power consumption.
A cloud system typically contains a huge number of computing nodes with numerous services running on. Nodes can fail in practice, leading to the downtime of services, hence, the reliability of cloud service is a cruci...
详细信息
ISBN:
(纸本)9781665423830
A cloud system typically contains a huge number of computing nodes with numerous services running on. Nodes can fail in practice, leading to the downtime of services, hence, the reliability of cloud service is a crucial issue for cloud service providers. To make the cloud environment more reliable, predicting the failure before it already happens is important. The ability to predict faulty nodes enables the migration of service to the healthy nodes, therefore improving service availability. Proactive fault prediction techniques with deep learning methods that are based on historical metrics (such as CPU usage, memory usage, etc) to predict future failures can be used to solve effectively this problem. In this paper, we propose to use the Bidirectional Long Short Term Memory (Bi-LSTM) model for predicting the failure-proneness of nodes in a cluster based on time series data.
With the advent of disruptive programmable network technologies such as SDN and NFV, network evolution has been spurred by accommodating cloud service infrastructure inside the network. Accordingly the key issues for ...
详细信息
ISBN:
(纸本)9781479967865
With the advent of disruptive programmable network technologies such as SDN and NFV, network evolution has been spurred by accommodating cloud service infrastructure inside the network. Accordingly the key issues for the service orchestration and harmonization to provide scalable and reliable application services on that environment would be investigated through the PoC service scenarios.
Since the late 1980s the world is working towards connectivity and convergence. In the last three decades, the convergence of information resources has happened. However to achieve a true convergence the information a...
详细信息
Cloud computingtechnology enables service to add special characteristics such as remote execution, service mobility and virtual workspace in a handheld device. However, it has a limitation on application areas for th...
详细信息
ISBN:
(纸本)9788988678558
Cloud computingtechnology enables service to add special characteristics such as remote execution, service mobility and virtual workspace in a handheld device. However, it has a limitation on application areas for the reason that mobile applications distinctly need a help of the host computer. To eliminate those threats, hash chain based approaches are proposed [1][2]. However, it also has some restriction that a mobile device sends request messages to each server when the device needs a multi-entity co-operative work. Therefore, this paper introduces a concept of secure entrustment, which provides most near and trusted servers whenever your mobile devices want to use a trusted computing.
Modern computer game engine has evolved to become a complete suite of virtual world constructing tools and runtime environment. The latter is usually a tightly integrated framework of 3D rendering engine, scripting en...
详细信息
ISBN:
(纸本)9780769534077
Modern computer game engine has evolved to become a complete suite of virtual world constructing tools and runtime environment. The latter is usually a tightly integrated framework of 3D rendering engine, scripting engine, physics simulation and networking. This framework design as well as individual component implementations decides the general type of games that could be composed by it. This article proposes a game engine framework called ParaEngine for developing games based on distributed game world data and logic. Its framework design is based on a modified version of Simulation Theory about the human brain. It has long been observed that our brain is both a distributed computing environment and a theatre of multimedia. The analogy of human cognition to simulation system has been applied to the proposed game engine to construct distributed Internet games. The implementation is illustrated in an RPG game demo called National Heroes.
In this paper, gives the definition of complex fuzzy matrix, and study its convergence problems based on the fuzzy matrix theory, which included the convergence in norm and the convergence in Power, some important con...
详细信息
ISBN:
(纸本)9783319191058;9783319191041
In this paper, gives the definition of complex fuzzy matrix, and study its convergence problems based on the fuzzy matrix theory, which included the convergence in norm and the convergence in Power, some important conclusions are obtained, to build and to improve the solid foundation for complex fuzzy matrix theory.
Artificial Intelligence is growing very fast. Along with the development of artificial intelligence, Ambient intelligence is also developing, which can identify and transmit various activities such as people's dai...
详细信息
ISBN:
(纸本)9781728108933
Artificial Intelligence is growing very fast. Along with the development of artificial intelligence, Ambient intelligence is also developing, which can identify and transmit various activities such as people's daily life and work. Ambient Intelligence has developed into convergencetechnology of Artificial Intelligence, Cloud computing and Internet of Things. We will discuss each technology. In this paper, we propose Personalized Ambient Intelligence which focuses on personalization based on the necessary element technologies for Ambient Intelligence(AmI). The proposed Personalized Ambient Intelligence should consist of Ambient Intelligence environment, AmI Common Data Model, Context Classification, AmI Sensor Technologies and Ambient intelligence situation. Among them, Personalized Ambient Situation should be classified into 4 situation as Home, Business, Driving, Moving. Using Personalized Ambient Intelligence presented in this paper, personalized context awareness is likely to be possible, and prediction of the next context will also be possible. And Finally, we will look at a brief summary of our definition of Personalized Ambient Intelligence and the areas that can be used in the future and conclude with future research.
Convolutional Neural Networks (CNNs) are currently the most advanced machine learning architecture for visual data classification. The choice of activation functions has a significant impact on the performance of a tr...
详细信息
ISBN:
(纸本)9781728103594
Convolutional Neural Networks (CNNs) are currently the most advanced machine learning architecture for visual data classification. The choice of activation functions has a significant impact on the performance of a training task. In order to overcome the vanishing gradient problem, we propose a new activation function for the classification system. The activation function makes use of the properties of periodic functions, where the derivative of a periodic function is also periodic. Furthermore, a linear combination is introduced to prevent the derivative from becoming zero. We verify this novel activation function by training an empirical analysis and comparing with the currently discovered activation functions. Experimental results show that our activation function SinP[N](x) = sin(x) + Nx, leads to very fast convergence even without the normalization layer. As a result, this new activation function enhances the training accuracy significantly, and can be easily deployed in the current systems built upon the standard CNN architecture.
Pervasive computing environment is high dynamic, and programs need to be enough intelligent to adapt to the dynamicity. Flexibility and adaptability are important features of programs in that environment. Our research...
详细信息
ISBN:
(纸本)9780769534077
Pervasive computing environment is high dynamic, and programs need to be enough intelligent to adapt to the dynamicity. Flexibility and adaptability are important features of programs in that environment. Our research team had proposed an Isotope Programming Model (IPM) to support developers in programming the environment. IPM mainly changes the traditional object-oriented approach that put all logic judgment of context in one method, and separates method from object of main program to isotope methods files by adding context block to each isotope one. Through computing the logic value in the context block using the given context, the system will choose the proper isotope method to execute. IPM promotes the adaptability of programs, but it is incomplete to some extent. In this paper, we extent IPM by utilizing ontology-based context model to separate attributes values from object. Object attributes values, parts of the context of environment, are modeled and stored in context ontologies database. An execution container will accomplish the matching procedure and return exact attributes values to object in programs. This makes the programs logic be more sensitive to changing of the environments. We also illustrate how to program in extended IPM (EIPM), and show that the separation will further promote flexibility and adaptability of programs.
暂无评论