No description available.

No description available.

No description available.

No description available.

No description available.

Recent technical advances have fuelled the popularity of mobile computing. Mobile devices such as smart phones and personal digital assistants are becoming more commonly used due to the reduction in their size and increase of computational power. In addition, wireless network hotspots (in airports, hotels and commercial outlets) are now beginning to populate the environment. With these advances, new types of mobile applications are becoming available to support users on the move. The mobile environment presents a number of challenges to application developers (including frequent network disconnection and variable bandwidth); therefore mobile middleware platforms have emerged to simplify the development process of distributed mobile applications. However, the range of platforms now available introduces the new problem of middleware heterogeneity, i.e., applications developed upon different types of middleware do not interoperate with one another. Hence, the next generation of mobile computing applications must be developed independently of specific middleware implementation to allow them to continue interoperating in new locations. This thesis investigates the problem of middleware heterogeneity in the mobile computing environment. The approach taken to solve this problem involves the development of a component-based, higher-level middleware framework (named ReMMoC) that can dynamically adapt its underlying behaviour between different concrete middleware implementations e.g. in one location CORBA is utilised, whereas at the next location SOAP is used. Furthermore, this framework promotes a higher-level programming abstraction based upon the abstract services concepts of the Web Services Architecture. The ReMMoC framework is evaluated to ensure that middleware transparency is achieved and that applications can be developed that will operate in unknown locations across unpredictable middleware implementation. Inevitably, the ability to overcome heterogeneity comes at the cost of an incurred performance overhead; hence, this thesis also evaluates the impact of this overhead in the domain of mobile computing.

No description available.

No description available.

New interactive computing applications are continually being developed in a bid to support people's changing work and recreational activities. As research focuses on one particular class of interactive systems, high level models of interaction are formulated and requirements emerge that reflect shared features or common functionality among those systems. The existence of models of interaction and shared functional requirements mean that support tools can be created which ease the subsequent development of these systems. Support tools most often take the form of architectures or frameworks that describe how a system should be structured. The type of tool that interactive systems developers are most familiar with is a library of reusable code that can be used for prototyping and building interactive applications and their interfaces. Within this thesis a new class of interactive system is identified, based on shared requirements for detection, processing and presentation of human physiological information. We have named these systems electrophysiologically interactive computer systems (EPICS) and describe in this thesis both the physiological and technological details behind their operation. A review is presented of existing research and development into this exciting new area of human-computer interaction, the aim being to establish the common requirements. These have enabled us to develop a suite of software components to support the creation of future EPIC systems. It is envisaged that the work presented in this thesis will serve as a jumping off point for others interested in exploring the potential of incorporating physiological information into the human-machine relationship.

No description available.