Improved adaptive signal power loss prediction using combined vector statistics based smoothing and neural network approach

Abstract

Predicting signal power loss between the transmitter and receiver with minimal error is an important issue in telecommunication network planning and optimization process. In recent years, median order statistic filters have been exploited as a preprocessing constituent for analyzing signals. This work presents a resourceful predictive model, built on multi-layer perceptron (MLP) network with vector order statistic filter based preprocessing technique for improved prediction of measured signal power loss in a microcellular LTE network environment. The predictive model is termed Vector statistic filters multilayer perceptron (VSF-MLP). In terms of some essential performance evaluation indices such as the correlation coefficient, root-mean-square error and coefficient of efficiency, results show that VSFMLP model prediction performs considerably better than the standard MLP model prediction approach on signal power data collected from different study locations in typical urban terrain.