Sub-optimal solutions for location and paging areas dimensioning in cellular networks

After a brief introduction to localization and paging concepts in cellular networks, this paper deals with the problem of optimum dimensioning for location areas (LA) and paging areas (PA) when they are coincident and time invariable. An analytical solution to this problem is first presented, even though this result can only be obtained with a strong simplification of the real case. Furthermore, this solution only provides a rough dimensioning for location areas. In the real case, where both cells and LAs have irregular boundaries, an analytical solution is not possible because of the complexity of the structure. This justifies that the LA dimensioning problem is solved by means of a heuristic method. It makes use of a self-adjusting algorithm that, with the support of a simulation model, modifies the localization structure of the cellular network reducing step by step a cost function until a value, reasonably close to the minimum value (sub-optimal solution) is achieved. The cost function takes into account the costs of a change of location area and the paging cost as well. The originality of the method consists in avoiding the approximation of cell and location area contours with simple geometrical figures. This ensures a much better dimensioning of the LAs with respect to the methods shown in the literature. Finally, the paper shows some application results of the self-adjusting method to a case study, namely the cellular network deployed over a metropolitan area. These results show both the potentials and the accuracy of the method