A random walk mobility model for location management in wireless networks

This paper describes a new and simple random walk mobility model that simplifies the two-dimensional Markov chain based on the properties of symmetry and lumped process. Compared to various other approaches, the proposed approach significantly reduces the complexity of the model by reducing the computing states. Based on this model, the location update rate and dwell time can be easily derived. The regular Markov chain and the property of combining states are applied to derive the number of location updates and a modified absorbing Markov chain property can be used to derive the dwell time. Analytical performance results of the model were validated by simulation. Results show the relative error between analytical and simulation performances are within 1%. This is the first model of its kind that can be used for studying area-crossing rates. This model can be adapted to most mobility management studies for architectures