ANN-Based Model for Predicting RF Signal Received Power in Indoor Propagation Links

Abstract

This chapter presents the use of Artificial Neural Networks (ANN) to predict the received power/path loss in indoor links. The prediction approach combines the use of ANNs and ray-tracing in order to identify and parameterize the so-called dominant path. A complete description of the process for creating and training an ANN-based model is presented, in which past values of one or more time series are used to predict future values. Special emphasis is placed on the training process. More specifically, we will be discussing various techniques to arrive at valid predictions focusing on an optimum selection of the training set. A quantitative analysis based on results from narrowband measurement campaigns is also presented.