Abstarct: In this thesis, our aim was to investigate the methods of pilot power optimization in communication cellular networks. The result is reducing in power consumption,that is one of the most important parameters of a radio system. We consider the problem of minimizing the total amount of pilot power subject to coverage constraint. This trade off apply in every cells and reduces total amount of pilot power consumption. Our goal is to design iterative algorithm for pilot power optimization subject to various coverage constraint levels, which allows us to generate a near optimal solution. Furthermore, to evaluate proposed methods, we used a test network originating from a planning scenario. Hence, we use an alghorithm baxsed on two main parts, one of which is aimed to solve a relaxed problem. In the second part, we adjust the current solution of the relaxed problem to a feasible solution that satisfies all the constraints of the enhanced model. The first part, which gives us a lower bound, is to be solved iteratively, within a chosen number of steps, independently on the second part, using a standard sungradient method. In every iteration, we solved a binary knapsack problem that indicates the areas that must be covered to guarantee a given degree of traffic coverage. We also compute a step size which specifies how far we move in a subgradient direction and then update the Lagrange multipliers. The numerical results demonstrate that the proposed method can give a reduction of up to 76% in the optimal pilot power levels compared to common pilot power assignment methods.