Neural network based optimal placement of base stations in three dimensional plane

Abstract

The paper proposes a solution to the problem of optimal placement of base stations, with effective utilization of resources. First, it analyzes the placement of base stations in x-y plane. Then this analysis is extended to 3-dimenional system, by including the z coordinate for representing height. This analysis optimizing these xy coordinate values as well as value of z coordinate, as height also contributes in covering the area of base stations. The Hopfield neural network model is used to find the solution to our problem.. Neural network and all parameters, i.e. x, y and z coordinates are used to maximize the coverage area and to minimize the interference, as optimal placement of base stations refer to these two points. In wireless communication system, the placement of base stations requires work force and costly equipments like GPS etc. resulting in loss of time and money. However, with proposed strategy only a laptop or desktop PC will be required to compute the positions of base stations within very less time. The energy equations are developed for our network which shows that the solution resulting from these equations is stable. The computer simulations and graphs are also included to validate the results.