Reasoning about uncertain contexts in pervasive computing environments

Context-aware systems can't always identify the current context precisely, so they need support for handling uncertainty. A prototype pervasive computing infrastructure, Gaia, allows applications and services to reason about uncertainty using mechanisms such as probabilistic logic, fuzzy logic, and Bayesian networks.     

Semantic management of multiple contexts in a pervasive computing framework

Mobile devices can perceive greater details of user states with the increasing integration of mobile sensors into a pervasive computing framework, yet they consume large amounts of batteries and computational resources. This paper proposes a semantic management method which efficiently integrates multiple contexts into the mobile system by analyzing the semantic hierarchy and temporal relations. The proposed method semantically decides the recognition order of the contexts and identifies each context using a corresponding dynamic Bayesian network (DBN). To sort out the contexts, we designed a semantic network using a knowledge-driven approach, whereas DBNs are constructed with a data-driven approach. The proposed method was validated on a pervasive computing framework, which included multiple mobile sensors (such as motion sensors, data-gloves, and bio-signal sensors). Experimental results showed that the semantic management of multiple contexts dramatically reduced the recognition cost.     

A simulation model for localization of pervasive objects using heterogeneous wireless networks

Position represents a relevant attribute needed by many applications, whose contexts are characterized by pervasiveness of the objects/things in the considered scenarios. In order to infer positions of pervasive objects, which are neither equipped with any location-sensing technologies, nor are unable to locate themselves, it is needed to support them with an appropriate infrastructure, which allows to determine their position in a manner that is transparent to applications. In this work, we aim at extending the positioning service using an agent-based approach, in order to discover and localize different kinds of objects, exploiting cheap and embedded technologies. We describe the design and the implementation of a layered architecture, that supports the localization of devices, and simple pervasive and ubiquitous objects. A simulation tool has been develop to evaluate the proposed solution in different application scenarios of every-day life.     

Experimenting with IDA* search algorithm in heterogeneous pervasive environments

Today, mobile and smart phones are often viewed as enablers of pervasive computing systems because they provide anytime and anywhere access to information services and computational resources. However, mobile devices are inherently constrained in their computational power and battery capacity making them mere “dumb terminals” connected to a resource-rich pervasive environment. If they are ever to play a more prominent role as true elements of a pervasive environment, mobile devices must be able to embed more application logic and delegate processing requests to pervasive infrastructure. In this paper we discuss distribution and offloading of computationally intensive tasks in pervasive environments populated by mobile devices. This approach is illustrated by experimenting with a distributed version of iterative deepening A* search algorithm. In our approach, the solution space of a problem being solved is partitioned and distributed among heterogeneous mobile devices, which yields a significant increase in the time of finding an optimal solution. Distributed IDA* search algorithm does not require any coordination or communication between mobile devices, but added inter-processor communication through shared memory further increases the efficiency of the algorithm. This paper presents the results of our experiments with the algorithm and discusses a number of issues related to its implementation.     

Enhancing pervasive Web accessibility with rule-based adaptation strategy

Content adaptation has shown significant benefits by enhancing Web accessibility to meet user’s requirements in heterogeneous environments. Adaptation rules can provide an efficient way to transform Web contents into adapted ones that conform to requesters’ contexts. However, the rule base management becomes a challenge while taking requesters’ contexts into account. In this paper, we refer requesters’ contextual requirements (or contexts in short) to the information that may affect their perceptions of browsing Web contents, such as requesters’ preferences and sensations, situated places, as well as devices and network connectivity. We have used JESS to design and implement dynamic adaptation strategies to guide the transformation process. Adaptation strategies are designed to improve rule base efficiency by performing dynamic scheduling of rule firing in an agenda. An adaptation strategy is composed of selected adaptation rules based on requesters’ contexts. When a user’s context changed, new adaptation strategy will be generated automatically by rearranging adaptation rules. By using JESS, we can dramatically reduce the number of adaptation rules from 54 to 22. The experiment results indicate our adaptation strategy can successfully adapt Web contents to satisfy requesters’ contexts and obtain superior performance.     

Implicit interaction design for pervasive workflows

The automatic identification capabilities of mobile devices enable the use of implicit interactions to connect the physical world with digital services. For example, users can be provided with information and services just by approaching to them or pointing to them with a mobile device. This kind of interactions can improve business processes by reducing the gap between physical and digital spaces. This work presents Parkour, a design method for workflows that make use of implicit interactions. Parkour allows designers to indicate how implicit interactions can be orchestrated to support a workflow. Furthermore, reconfiguration techniques have been applied to adapt at run-time the degree in which these interactions intrude the user’s mind. Tool support has been developed to automate the validation of the workflow models and support the system reconfiguration. Finally, the proposal has been applied in the development of several workflows.     

Enabling run-time composition and support for heterogeneous pervasive multi-agent systems

User needs-driven and computer-supported development of pervasive heterogeneous and dynamic multi-agent systems remains a great challenge for agent research community. This paper presents an innovative approach to composing, validating and supporting multi-agent systems at run-time. Multi-agent systems (MASs) can and should be assembled quasi-automatically and dynamically based on high-level user specifications which are transformed into a shared and common goal–mission. Dynamically generating agents could also be supported as a pervasive service. Heterogeneity of MASs refers to diverse functionality and constituency of the system which include mobile as well as host associated software agents. This paper proposes and demonstrates on-demand and just-in-time agent composition approach which is combined with run-time support for MASs. Run-time support is based on mission cost-efficiency and shared objectives which enable termination, generation, injection and replacement of software agents as the mission evolves at run-time. We present the formal underpinning of our approach and describe the prototype tool – called eHermes, which has been implemented using available agent platforms. Analysis and results of evaluating eHermes are presented and discussed.     

On enabling integrated process compliance with semantic constraints in process management systems

Key to broad use of process management systems (PrMS) in practice is their ability to foster and ease the implementation, execution, monitoring, and adaptation of business processes while still being able to ensure robust and error-free process enactment. To meet these demands a variety of mechanisms has been developed to prevent errors at the structural level (e.g., deadlocks). In many application domains, however, processes often have to comply with business level rules and policies (i.e., semantic constraints) as well. Hence, to ensure error-free executions at the semantic level, PrMS need certain control mechanisms for validating and ensuring the compliance with semantic constraints. In this paper, we discuss fundamental requirements for a comprehensive support of semantic constraints in PrMS. Moreover, we provide a survey on existing approaches and discuss to what extent they are able to meet the requirements and which challenges still have to be tackled. In order to tackle the particular challenge of providing integrated compliance support over the process lifecycle, we introduce the SeaFlows framework. The framework introduces a behavioural level view on processes which serves a conceptual process representation for constraint specification approaches. Further, it provides general compliance criteria for static compliance validation but also for dealing with process changes. Altogether, the SeaFlows framework can serve as formal basis for realizing integrated support of semantic constraints in PrMS.     

Aspect-oriented design and implementation of context-aware pervasive applications

Pervasive applications must be aware of the contexts where they are executed. These contexts may vary greatly and change quickly. Two main problems are associated with this issue: (1) context-awareness is a crosscutting concern that cannot be well-encapsulated in a single module using traditional technologies, thus hindering software maintenance and reusability; and (2) reasoning about application design correctness can be complex due to the number and diversity of potential contexts where a pervasive application could be executed. In order to overcome these problems, we present a process for the design and implementation of context-aware pervasive applications that uses aspect-orientation and executable modelling in order to overcome these shortcomings. Aspect-oriented techniques contribute to the encapsulation of crosscutting concerns, such as context-awareness, into well-localized modules. Executable modelling helps engineers to reason about application design by executing the design models in different contexts and situations. Pervasive applications are modelled using the aspect-oriented executable modelling UML 2.0 profile, executed at the modelling level for testing purposes, and then mapped into an aspect-oriented middleware platform for pervasive applications. This process is illustrated using a location-aware intelligent transportation system consisting of a set of cooperating sentient vehicles. This work has been supported by the Spanish Ministerio de Ciencia y Tecnologa (MCYT) Project TIN2005-09405-C02-01 and the European Commission Grant IST-2-004349-NOE AOSD-Europe and the European Commission STREP Project AMPLE IST-033710.     

Cross-dataset heterogeneous adaptation learning based facial attributes estimation

Multimedia Tools and Applications - Recently, human facial attributes analysis has become an important research topic in the field of pattern recognition and computer vision. In fact, various tasks...     

异构感知数据的动态适配接入方法

开放地理空间信息联盟的传感器网整合框架（ SWE）定义了异构传感器统一描述标准,却未定义底层的异构数据接入该框架的过程,导致不同类型的感知数据接入时需要人工提供不同的数据接入方式。针对以上问题,设计了异构传感器信息模型（ HSIM）,描述了传感器的自身属性、观测能力等特征,同时概括了传感器数据结构和数据传输协议的特征;提出了一种异构感知数据动态适配接入方法,该方法通过动态生成适配器实现异构感知数据的自动解析和接入SWE的过程。气象异构数据适配接入系统表明：该方法能够在较少的人工干预下实现物理传感器与虚拟传感器异构感知数据的自动解析与接入,为异构感知数据动态集成提供了高效率的解决方案。     

A model of domain-polymorph component for heterogeneous system design

Heterogeneous systems mix different technical domains such as signal processing, analog and digital electronics, software, telecommunication protocols, etc. Heterogeneous systems are composed of subsystems that are designed using different models of computation (MoC). These MoCs are the laws that govern the interactions of the components of a subsystem. The design of heterogeneous systems includes the design of each part of the system according to its specific MoC, and the connection of the parts in order to build the model representing the system. Indeed, this model allows the MoCs that govern different parts of system to coexist and interact.To be able to use a component which is specified according to a given MoC, under other, different MoCs, we can use either a hierarchical or a non-hierarchical approach, or we can build domain-specific components (DSC). However, these solutions present several disadvantages. This paper presents a new model of component, called domain-polymorph component (DPC). Such a component is atomic and is able to execute its core behavior, specified under a given MoC, under different host MoCs. This approach is not a competitor to the approaches above but is complementary.     

A B-spline-based approach to heterogeneous objects design and analysis

The recent advancement of solid freeform fabrication, design techniques and fundamental understanding of material properties in functionally graded materials has made it possible to design and fabricate multifunctional heterogeneous objects. In this paper, we present an integrated design and analysis approach for heterogeneous object realization, which employs a unified design and analysis model based on B-spline representation and allows for direct interaction between the design and analysis model without laborious meshing operation. In the design module, a new approach for intuitively modelling of multi-material objects, termed heterogeneous lofting, is presented. In the analysis module, a novel graded B-spline finite element solution procedure is described, which gives orders of magnitude of better convergence rate in comparison with current methods, as demonstrated in several case studies. Further advantages of this approach include simplified mesh construction, exact geometry/material composition representation and easy extraction of an isomaterial surface for manufacturing process planning.     

Peer-assisted adaptation in periodic broadcasting of videos for heterogeneous clients

The increasing use of streaming applications and wide variety of devices, such as handheld personal organizers, have raised new challenges, such as the provision of access to multimedia content for devices with limited computing and bandwidth capabilities. In this paper, we propose a technique for periodic broadcasting. Our technique allows clients with low bandwidth capability to be served in the same period as high-bandwidth clients, using peer-to-peer streaming between clients, and also works for clients with downstream capability below the video bit rate. In the proposed system, the server builds multicast trees of clients that allow clients with low bandwidth to be served by other clients with sufficient upstream bandwidth. Clients below the video bit rate will suffer additional, but only moderately noticeable, delays that occur as video is converted to lower quality. We performed extensive simulation and showed that our system can provide improved service performance for heterogeneous clients on periodic broadcasting protocols.     

Feature-based design for heterogeneous objects

Heterogeneous objects are objects composed of different constituent materials. In these objects, multiple desirable properties from different constituent materials can be synthesized into one part. In order to obtain mass applications of such heterogeneous objects, efficient and effective design methodologies for heterogeneous objects are crucial.In this paper, we present a feature based design methodology to facilitate heterogeneous object design. Under this methodology, designers design heterogeneous objects using high-level design components that have engineering significance. These high level components are form features and material features. In this paper, we first examine the relationships between form features and material features in heterogeneous objects. We then propose three synthesized material features in accordance with our examination of these features. Based on these proposed features, we develop a feature based design methodology for heterogeneous objects. Two enabling methods for this design methodology, material heterogeneity specification within each feature and combination of these material features, are developed. A physics (diffusion) based B-spline method is developed to (1) allow design intent of material variation be explicitly captured by boundary conditions, (2) ensure smooth material variation across the feature volume. A novel method, direct face neighborhood alteration, is developed to increase the efficiency of combining heterogeneous material features.Examples of using this feature based design methodology for heterogeneous object design, such as a prosthesis design, are presented.     

普适计算安全体系结构模型研究与设计

根据目前国际上普适计算研究的现状和发展趋势--国际上对普适计算的研究主要集中在人机接口和上下文感知计算等领域,国内对普适计算的研究主要集中在多模态的智能空间领域.提出了一个普适计算安全体系结构的参考模型,并简要讨论了该参考模型的3个关键研究内容--安全系统层、安全计算层和安全协同层.最后列举了一个模型实例.