Environment-Adaptation Based Hybrid Neural Network Predictor for Signal Propagation Loss Prediction in Cluttered and Open Urban Microcells

Abstract

In the design and placement of radio base station transmitters, the accurate field signal power prediction and modelling is of critical importance. In this work, an adaptive neural network predictor which combines multilayer perception (MLP) and adaptive linear element (Adaline) is proposed for enhanced signal propagation loss prediction in microcellular urban environments. The prediction accuracy of the proposed Hybrid adaptive neural network predictor has been tested and evaluated using experimental field strength data acquired from LTE radio network environment with mixed residential, commercial and cluttered building structures. By means of first order statistical performance evaluation metrics, namely, regression coefficient (R), root mean squared error, standard deviation and mean absolute error, the proposed adaptive hybrid approach provide a better prediction accuracy compared to the standard MLP ANN prediction approach. The superior performance of the hybrid neural predictor can be attributed to its capability to learn, adaptively respond and predict the fluctuating patterns of the reference propagation loss data during training.