Location uncertainty in mobile networks: a theoretical framework

Abstract

It is difficult to compare various mobility management schemes since as yet there exist no absolute bounds on optimum performance which can be applied to any procedure regardless of assumptions about mobility, network structure, and the like. It would be useful to fill this gap by developing some general theory, which optimally would precisely relate quantities of interest such as minimum average effort/delay in determining unit locations and minimum average location information dissemination rate over the network in fundamental terms. We therefore note that in order to quantitatively evaluate the average cost of any mobility management scheme, timevarying probability distributions on mobile unit location are required. These distributions might be derived from specific motion models, approximated via empirical data [18, 33], or

Performance Evaluation of an Enhanced Distance-Based Registration Scheme Using the Normal Distribution Approximation

“…Several mobility models have been proposed; a simple onedimensional model [2], a random walk model [3,[7][8], a fluid flow model [9], the Brownian motion model [10], and a four-directional model [4][5]. Among them, a simple onedimensional model is too simple to be applied for exact analysis.…”

“…Many schemes have been proposed in order to minimize the location registration cost [1][2][3][4][5][6][7][8][9][10]. In particular, nine different registration schemes have been recommended in CDMA mobile cellular systems [1] as shown in Table 1.…”

Analysis Of the Paging Costs In Location Management for Mobile Communications Networks Leveraging The Diffusion Constant

“…This strategy is especially suitable when the mobile terminal changes location according to stochastic processes. Considering for example an isotropic Brownian motion process with drift, the Gaussian probability density function for the location of the mobile can be applied [18], and the location probability can be obtained through the integration of the density function over the region of interest. For the particular analysis of the paging costs, an approach based on location probability distributions could be the information theoretical one, by means of relating the entropy of the distribution with the cost of paging, as the mean of the ordered distribution corresponds to the minimum paging costs [19].…”

GPS-free positioning in mobile ad-hoc networks

“…In [6] Rose and Yates give a theoretical framework of the mobility and location tracking in mobile systems. They present a study of mobility tracking based on user/service/host location probability distribution.…”