LTE Network Planning and Traffic Generation

The main objective of this thesis is to propose automatic planning tools for the planning problem of Long Term Evolution (LTE) networks based on realistic traffic. More precisely, a set of equations is first proposed to generate realistic traffic profile while considering as many practical aspects as possible. Then, a mathematical model is created to solve the planning problem of the evolved packet core of LTE networks while minimizing the cost. Since the planning problem is NP-Hard, an approximate method based on the local search is also proposed. The algorithm provides a trade-off between the solution quality and the execution time. Numerical results show that the approximate approach is capable of providing good solutions (on average within 4.03% of the optimal solution) in shorter time duration than the exact approach. ii Acknowledgements