ITC/USA 2007 Conference Proceedings / The Forty-Third Annual International Telemetering Conference and Technical Exhibition / October 22-25, 2007 / Riviera Hotel & Convention Center, Las Vegas, Nevada / Increasingly, many software developers are facing the challenge of adapting software applications developed on one platform to work on multiple platforms. While software standards have helped this effort, they do not go far enough, and many platforms only partially support these standards leaving many needed functions in platform specific libraries. This is particularly evident in the areas of graphics and user interfaces, threading and synchronization, and in network and file access. Fortunately, Java offers a common interface where native libraries diverge. This paper outlines a phased strategy for migrating platform specific applications to be platform independent while reusing the robust, existing algorithms.

The widespread deployment of inexpensive communications technology, computational resources in the networking infrastructure and network-enabled end devices pose a problem for end users: how to locate a particular network service or device out of those accessible. Service providers use Service Discovery Services (SDS) to advertise the descriptions of available or already running services, while clients use SDS to compose queries for locating these services. Service descriptions and queries use the eXtensible Markup Language (XML) to encode vendor specific information and device- or service-specific capabilities as well as the actions addressed to the device or service. This report presents the architecture and implementation of a SDS used to locate enabled LCD projectors and use them for presentation. The presentation service provides all the capabilities to the end user so that he can choose the projector device of his interest and use the graphical user interface to navigate thorough the presentation. The presentation service also has the capability to use more than one projector at a time. We use the Universal Plug and Play suite of protocols to establish the communication between client and the projector device.

Conventionally, pilots control ROV (Remotely Operated Vehicle) with on-board console connected to the vehicle with a power/signal tether. For those who want to operate the vehicle or observe the underwater scenery would need to sit right by the console. Operation of the vehicle or the real time data acquired by vehicle's sensors is not possible or available for the scientist who is not on-board. Recently, there has been a vast development and application of Internet technology on both new emerging fields and conventional fields as well. This motivates us to design an Internet-based man-machine interface to overcome the aforementioned limits. We choose Java to be the core language for its excellency in communication, security, portability and graphics user interface (GUI). On the other hand, Java suffers from lacking the interface to do low level access to the hardware of the computer. This means that Java can not accomplish hardware control by itself. However, this limitation can be removed by including JNI (Java Native Interface) which provides a protocol for Java to communicate with DLL written in C. Security is also a major concern of Internet-based applications. We use Java's built-in security package to carry out "Message Digest" computation for the login authentication. Different level of privileges are given to the users according to their user name and password. So any user who has access to Internet can use internet browser (Microsoft IE or Netscape) to control the system remotely. In this thesis, we use a CCD camera mounted on a stepping motor to represent the whole ROV system. Different subject users over the TANET, including NTOU in Keelung, ME and IUT NSYSU in Kaohsiung, are tested. We find that under the current internet infrastructure in Taiwan, our system can deliver as least one 160x120 24-bit color frame every 2 seconds or so. Control commands for the stepping motor experiences no significant delay. This preliminary result indicates a similar structure can be adopted in developing I.A (Information Appliance).

Java est à ce jour l'un des langages, si ce n'est le langage, le plus utilisé toutes catégories de programmation confondues et sa popularité concernant le développement d'applications scientifiques n'est plus à démontrer. Néanmoins son utilisation dans le domaine du Calcul Haute Performance (HPC) reste marginale même si elle s'inscrit au cœur de la stratégie de certaine entreprise comme Aselta Nanographics, éditeur de l'application Inscale pour la modélisation des processus de lithographie par faisceaux d'électron, instigateur et partenaire industriel de cette thèse.Et pour cause, sa définition haut-niveau et machine-indépendante, reposant sur un environnement d'exécution, parait peu compatible avec le besoin de contrôle bas-niveau nécessaire pour exploiter de manière optimale des architectures de microprocesseurs de plus en plus complexes comme les architectures Intel64 (implémentation Intel de l'architecture x86-64).Cette responsabilité est entièrement déléguée à l'environnement d'exécution, notamment par le biais de la compilation dynamique, chargée de générer du code binaire applicatif à la volée. C'est le cas de la JVM HotSpot, au centre de cette étude, qui s'est imposée comme l'environnement de référence pour l'exécution d'applications Java en production.Cette thèse propose, dans ce contexte, de répondre à la problématique suivante : comment optimiser les performances de code séquentiel Java plus particulièrement dans un environnement HotSpot/Intel64 ?Pour tenter d'y répondre, trois axes principaux ont été explorés. Le premier axe est l'analyse des performances du polymorphisme, mécanisme Java haut-niveau omniprésent dans les applications, dans le lequel on tente de mesurer l'impact du polymorphisme sur les performances du code et d'évaluer des alternatives possibles. Le second axe est l'intégration de code natif au sein des applications - afin de bénéficier d'optimisations natives - avec prise en compte du compromis coût d'intégration/qualité du code. Enfin le troisième axe est l'extension du compilateur dynamique pour des méthodes applicatives afin, là encore, de bénéficier d'optimisations natives tout en s'affranchissant du surcout inhérent à l'intégration de code natif.Ces trois axes couvrent différentes pistes exploitables dans un contexte de production qui doit intégrer certaines contraintes comme le temps de développement ou encore la maintenabilité du code. Ces pistes ont permis d'obtenir des gains de performances significatifs sur des sections de code applicatif qui demeuraient jusqu'alors très critiques. / L'auteur n'a pas fourni de résumé en anglais

No description available.

Syftet med denna studie är utforska möjligheterna att återanvända modellkod, skriven i C, över flera plattformar, nämligen iOS, Android och HTML5. Målet är att undersöka kodstorlek och komplexitet vid återanvändning av modellkod skriven i C på plattformarna. Genom en noggrann granskning av tidigare forskning och existerande metoder identifierades lämpliga verktyg och metoder. Det som framkom är att WebAssembly var det rekommenderade verktyget för att koppla C- kod till HTML5. För Android var Java Native Interface (JNI) det mest lämpliga verktyget för att koppla C-kod till Kotlin. Swift, å andra sidan, har inbyggda verktyg för interoperabilitet som möjliggör koppling mellan Swift-kod och C. Därav påvisade resultaten att det är möjligt att återanvända modellkod skriven i C på Android, iOS och HTML5. Dock, för Android krävdes det tillsats av extra kod för att hantera kopplingen mellan Kotlin och C-modellkod. Eftersom dessa språk hanterar datatyper och minneshantering på olika sätt. Inom iOS krävdes inte lika mycket extra kod som på Android för att hantera kopplingen mellan Swift och C eftersom Swift är utvecklat med naturligt stöd för C. I HTML5 behövdes inte lika mycket extra kod som på Swift och Android eftersom Emscripten möjliggjorde automatisk generering av en omvandling från modellkod skriven i C till JavaScript. Detta underlättade utvecklandet av bindningen mellan C-baserad modellkod och HTML5-applikationen. / The purpose of this study was to explore the possibilities of reusing model code, written in C, across multiple platforms, specifically iOS, Android, and HTML5. The objective was to investigate code size and complexity when reusing model code written in C on these platforms. Through a thorough review of previous research and existing methods, suitable tools and methods were identified. It was found that WebAssembly was the recommended tool for linking C code to HTML5. For Android, the Java Native Interface (JNI) was the most appropriate tool for linking C code to Kotlin. Swift, on the other hand, has built-in tools for interoperability that allow for the connection between Swift code and C. The results indicated that it is possible to reuse model code written in C on Android, iOS, and HTML5. For Android, additional code was required to manage the connection between Kotlin and C model code, as these languages handle data types and memory management differently. However, within iOS, there was not needed additional code compared to Android, to manage the linkage between Swift and C. Since Swift is developed with inherent support for C. For HTML5, there was not as much additional code required as in Swift and Android, primarily because Emscripten enabled automatic conversion of model code written in C to JavaScript. This facilitated the development of the binding between C-based model code and the HTML5 application.

In the past decade, the technologies of computer communication and PC hardware/software evolve in a very fast pace. Many conventional industries are refurbished with these new tools. Hence the operation of the industry can be improved, and even the products, equipped with new technology, demonstrate a new dimension of their function. Among all these new ideas or tools, we would like to study the feasibility of integrating WWW server, Java JNI, Real-time Linux and PC-based hardware components together to form an internet-based manufactory service server. In this work, we use Linux Redhat OS as the platform, and its Apache homepage server to provide users to access the services, such as activating a relay or retrieving the status of a limit switch. Enhanced with Java JNI, the WWW server can access the control of PC's hardware. More importantly, patching the OS with real-time packages, the WWW server is transformed into real-time controller which is much cheaper and much reliable than its opponents, such as Microsoft NT. RS 232 serial ports, an 8255 multi-function I/O card, optical encoder card, D/A card and a servo motor are integrated under the command of the WWW server. By browsing the control center's homepage, users can easily acquire the status of the peripherals, or send out control command remotely. Experiment results confirm the performance of the system. The structure of this experimental server can be modified to fit the requirement of a remotely operated or tele-monitoring system by rewriting the homepage.

Something that has recently gained popularity in the leading car manufacturing companies is the integration of an auto-reverse assistance system to improve the customer experience. The undeniable spread of smartphones and their significant role in human life in recent years, gave rise to the idea of designing an application to be used for reverse driving. As a result, the car company Volvo proposed an idea for this project: to implement an Android based application to facilitate reverse navigation in their trucks. From a technical point of view, the most crucial obstacle that should be addressed is implementing a secure and reliable communication link between the smartphone and the truck's control centre. Hence, the primary goal of this thesis work is to provide a secure channel to transmit data and computing reliability of the com-munication link. Another objective of this project is to define a solution for auto-converting path plan function, currently developed in Matlab, to run on Android devices. In addition, this solution should enable the developer to modify the path plan function in Matlab without having to consider Android programming. The differential method has been proposed for transferring secure data transfer using Bluetooth technol-ogy. This solution not only increases the security of the communication link but also improves transmission time. Another objective has been reached by developing a middleware function using C programming language and Android Native Development Kite (NDK) between An-droid and Matlab. The result of these solutions provides a reliable communication link and runs the developed path plan function in Android phone.

This thesis deals with processing of video content on Android platform. Libraries for video encoding and decoding usable on this platform are described first. One of these libraries - FFmpeg - is used as core to design and implement middleware video-processing library. This library will provide simple but e fective API for multimedia Java applications.

No description available.