|
|
Supporting User Mobility in Ubiquitous Computing Environment (Aug. 2001 - May. 2002) | |||||
|
Mobile Computing has been an active research area over the years. It has gained more intensive consideration
since the recent proliferation of network-enabled portable devices and inexpensive wireless communication. All these advancing
technologies are enabling "ubiquitous personal computing
environment". In such an environment, a user can access his/her personalized services at anytime, anywhere, through any possible
mobile/fixed terminals in a secure way. A wealth of research work has been conducted to approach this goal.
Most of them focused on the host mobility problem. With host mobility, a user can access the same service while
traveling. However, he/she is required to always carry the mobile host. We argue that this is only a special case of
user mobility. In the ubiquitous computing environment, user mobility} not only includes host mobility, but also includes the
case where a user is free to switch from one host to another. In a broader sense, a user is free to migrate his/her
service environment. This means that the user, instead of host, is the real end receiver of any service.
One fundamental problem is how to dynamically migrate the user's session as the user moves, so that the service always follows him/her. We refer to this problem as user-level handoff. This is in contrast to the traditional host-level handoff scenario, where a mobile host switches from one network to another. The unique challenges of this problem are as follows:
These problems demand mobility support from both system (OS, network) level and application level. However, supporting user mobility at application layer is not a viable solution. Making each application to be mobility-aware will greatly complicate its development. Moreover, legacy applications are not considered. On the other hand, system-level solutions are insufficient since they are not aware of application semantics. This is quite essential to address the problem of user-level handoff across heterogeneous platforms. We believe that the middleware solution is the best to address these problems due to its following advantages.
We designed and implemented a middleware framework to support user mobility in the Active Space of Gaia Project. Its major mobility functions include user-level handoff management and service instantiation across heterogeneous computing platforms. Furthermore, we make several enhancements to improve system performance, such as accelerating user-level handoff, therefore saving state information storage/transportation overhead. We tested our framework using two multimedia applications (Mobile Video Player and Mobile Audio Player).
Active Space
Screenshots of Mobile Video Player (Session Migration from desktop PC to iPaq)
| ||||||
Publication:
|