Skip to Main Content
In this paper the model of signal strength prediction that seeks an optimal location of single access point in indoor environment is presented. The location of the transmitter needs to ensure maximally uniform coverage with electromagnetic signal in the given environment. The propagation predictions are obtained by the model based on neural network. The advantage of the neural network model is given by their possibility to obtain enough accurate values of field strengths inside buildings without any knowledge about building geometry and construction characteristics. The neural network needs to be trained by measured signal strengths at chosen receiving locations. The model divides coverage environment into a grid of possible receiver locations and specifies cost function as a sum of weighted signal strengths for particular receiver points. This is nonlinear programming problem for minimizing the differences between given threshold signal strength and actual signal strength at particular location. As optimization method Particle Swarm Optimization (PSO) technique is used. The final results show good coverage with adequate signal strengths for entire given indoor environment.