Abstract: This paper discusses the improvements and parameter optimization
issues of LTRACK, a recently proposed mobility management algorithm.
Mathematical modelling of the algorithm and the behavior of the Mobile Node
(MN) are used to optimize the parameters of LTRACK. A numerical method is given
to determine the optimal values of the parameters. Markov chains are used to
model both the base algorithm and the so-called loop removal effect. An extended qualitative and quantitative analysis is carried out to
compare LTRACK to existing handover mechanisms such as MIP, Hierarchical Mobile
IP (HMIP), Dynamic Hierarchical Mobility Management Strategy (DHMIP),
Telecommunication Enhanced Mobile IP (TeleMIP), Cellular IP (CIP) and HAWAII.
LTRACK is sensitive to network topology and MN behavior so MN movement
modelling is also introduced and discussed with different topologies. The techniques presented here can not only be used to model the
LTRACK algorithm, but other algorithms too. There are many discussions and
calculations to support our mathematical model to prove that it is adequate in
many cases. The model is valid on various network levels, scalable vertically
in the ISO-OSI layers and also scales well with the number of network
elements.