Comparative Analysis of Basic Models and Artificial Neural Network Based Model for Path Loss Prediction

Abstract

Abstract-Propagation path loss models are useful for the prediction of received signal strength at a given distance from the transmitter; estimation of radio coverage areas of Base Transceiver Stations (BTS); frequency assignments; interference analysis; handover optimisation; and power level adjustments. Due to the differences in: environmental structures; local terrain profiles; and weather conditions, path loss prediction model for a given environment using any of the existing basic empirical models such as the Okumura-Hata's model has been shown to differ from the optimal empirical model appropriate for such an environment. In this paper, propagation parameters, such as distance between transmitting and receiving antennas, transmitting power and terrain elevation, using sea level as reference point, were used as inputs to Artificial Neural Network (ANN) for the development of an ANN based path loss model. Data were acquired in a drive test through selected rural and suburban routes in Minna and environs as dataset required for training ANN model. Multilayer perceptron (MLP) network parameters were varied during the performance evaluation process, and the weight and bias values of the best performed MLP network were extracted for the development of the ANN based path loss models for two different routes, namely rural and suburban routes. The performance of the developed ANN based path loss model was compared with some of the existing techniques and modified techniques. Using Root Mean Square Error (RMSE) obtained between the measured and the model outputs as a measure of performance, the newly developed ANN based path loss model performed better than the basic empirical path loss models considered such as: Hata; Egli; COST-231; Ericsson models and modified path loss approach.

Application of Computational Intelligence Algorithms in Radio Propagation: A Systematic Review and Metadata Analysis

“…Nadir and Idrees Ahmad [42] To address the applicability of the Okumura-Hata model in GSM frequency band of 890-960 MHz ANN Delos Angeles and Dadios [43] To predict path loss for TV transmission using alternative neural networks, and ascertain the proposed model viability ANN Benmus et al [44] To predict the propagation path loss with an empirical model at the capital city of Libya ANN Ofure et al [33] To use a three-stage approach in the determination of GSM Rx level from atmospheric parameters ANN Eichie et al [45] To develop an ANN-based path loss estimation model for rural and urban areas ANN…”

“…e performance of the proposed models was tested and found to be efficient with a mean squared error (MSE) of 0.056. Eichie et al in [45] developed an ANN-based path loss model for signal quality estimation in rural and urban areas. ey relied on a feedforward network topology and the Matlab Network Toolbox learning algorithm; specifically, they used the Levenberg-Marquardt approach with 31 to 39 neurons in incremental steps of 2.…”

Cellular Communications Coverage Prediction Techniques: A Survey and Comparison

“…The model applies to various scenarios without the accuracy been altered. Unfortunately, the process is quite complicated and lacks computational efficiency [138]. Some of the known and widely used deterministic models for the prediction of network coverage area are discussed.…”

Measurement Validation of Fuzzy-like Propagation Models for Wireless Cellular Networks in Metropolitan Environment

“…Fuzzy logic is used to determine the unknown environment, which is obtained using linguistic rules that provide a fine tuning of the known propagation environments. The authors in [8] propose some modifications to the COST 231 WI model to reflect the information related to building and terrain heights of practical environments. A dual slope propagation model is developed based on the modified WI model for small cells with electrical and mechanical tilt.…”