ABSTRACT
Apparently, from a network-technology-centric point of view the first — and only — thing moving is a
network
terminal
device
. Research in mobile computing should therefore focus on allowing mobile devices to easily disconnect and reconnect to different networks, but this seems rather insufficient. Because most terminal devices are not able to move autonomously, one could argue that research should rather have the
user
moving the device to be addressed. Additionally, such user mobility can be extended to cases where users move from one device to another, demanding support for the same services. This can be expressed with the more general term of “personal mobility.” Most users, however, do not care about the inner workings of their devices; they just want, among other things, the appropriate
services
to be available at all times, which we can denote as “service mobility.” Because a service is expressed through
program
code
, we have now identified the following four key types of mobility:
Personal mobility
Service mobility
Code mobility
Terminal mobility
, then, subsumes in general all changes in the network topology. These changes can be due to the physical disconnection of devices (and reconnection to other networks) or moving wireless network devices in and out of the radio footprint of their neighbors. Examples are laptops connected via Ethernet or wireless LAN or mobile phones moving from one base station to another. Coping with terminal mobility is a well-known and well-addressed problem, as the network reference model implied that providing network access was the first (and to some people only) problem to solve; however, research in mobile computing showed that this approach must be extended for most practical applications of mobile users.
