Artificial Neural Network Optimal Modeling and Optimization of UAV Measurements for Mobile Communications Using the L-SHADE Algorithm

Channel modeling of wireless communications from unmanned aerial vehicles (UAVs) is an emerging research challenge. In this paper, we propose a solution to this issue by applying a new framework for the prediction of received signal strength (RSS) in mobile communications based on artificial neural networks (ANNs). The experimental data measurements are taken with a UAV at different altitudes. We apply several evolutionary algorithms (EAs) in conjunction with the Levenberg–Marquardt (LM) backpropagation algorithm in order to train different ANNs and in particular the L-SHADE algorithm, which self-adapts control parameters and dynamically adjusts population size. Five new hybrid training methods are designed by combining LM with self-adaptive differential evolution (DE) strategies. These new training methods obtain better performance to ANN weight optimization than the original LM method. The received results are compared with the real values using representative ANN performance indices and exhibit satisfactory accuracy.

[1]  A. Massa,et al.  Bayesian Compressive Sampling for Pattern Synthesis With Maximally Sparse Non-Uniform Linear Arrays , 2011, IEEE Transactions on Antennas and Propagation.

[2]  Arthur C. Sanderson,et al.  JADE: Adaptive Differential Evolution With Optional External Archive , 2009, IEEE Transactions on Evolutionary Computation.

[3]  Halim Yanikomeroglu,et al.  The New Frontier in RAN Heterogeneity: Multi-Tier Drone-Cells , 2016, IEEE Communications Magazine.

[4]  George V. Tsoulos,et al.  Path Loss characteristics for UAV-to-Ground Wireless Channels , 2019, 2019 13th European Conference on Antennas and Propagation (EuCAP).

[5]  Theodore S. Rappaport,et al.  Wireless communications - principles and practice , 1996 .

[6]  Jan Prokopec,et al.  Propagation path loss models for mobile communication , 2011, Proceedings of 21st International Conference Radioelektronika 2011.

[7]  J. N. Sahalos,et al.  Modeling by optimal Artificial Neural Networks the prediction of propagation path loss in urban environments , 2013, 2013 IEEE-APS Topical Conference on Antennas and Propagation in Wireless Communications (APWC).

[8]  Jing Liu,et al.  A Multi-Objective Evolutionary Algorithm for Promoting the Emergence of Cooperation and Controllable Robustness on Directed Networks , 2018, IEEE Transactions on Network Science and Engineering.

[9]  Qingfu Zhang,et al.  Differential Evolution With Composite Trial Vector Generation Strategies and Control Parameters , 2011, IEEE Transactions on Evolutionary Computation.

[10]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[11]  P. N. Suganthan,et al.  Differential Evolution Algorithm With Strategy Adaptation for Global Numerical Optimization , 2009, IEEE Transactions on Evolutionary Computation.

[12]  P. Rocca,et al.  Differential Evolution as Applied to Electromagnetics , 2011, IEEE Antennas and Propagation Magazine.

[13]  Janez Brest,et al.  Self-Adapting Control Parameters in Differential Evolution: A Comparative Study on Numerical Benchmark Problems , 2006, IEEE Transactions on Evolutionary Computation.

[14]  Michael Georgiopoulos,et al.  Applications of Neural Networks in Electromagnetics , 2001 .

[15]  Dimitra Zarbouti,et al.  LTE Ground-to-air Measurements for UAV-assisted cellular networks , 2018 .

[16]  Jeroen Wigard,et al.  Radio Channel Modeling for UAV Communication Over Cellular Networks , 2017, IEEE Wireless Communications Letters.

[17]  A. Massa,et al.  Optimization of the Directivity of a Monopulse Antenna With a Subarray Weighting by a Hybrid Differential Evolution Method , 2006, IEEE Antennas and Wireless Propagation Letters.

[18]  George V. Tsoulos,et al.  LTE Ground-to-Air Field Measurements in the Context of Flying Relays , 2019, IEEE Wireless Communications.

[19]  Bidyadhar Subudhi,et al.  Differential Evolution and Levenberg Marquardt Trained Neural Network Scheme for Nonlinear System Identification , 2008, Neural Processing Letters.

[20]  Raj Mittra,et al.  Evolutionary Algorithms Applied to Antennas and Propagation: A Review of State of the Art , 2016 .

[21]  A. Massa,et al.  Optimization of the difference patterns for monopulse antennas by a hybrid real/integer-coded differential evolution method , 2005, IEEE Transactions on Antennas and Propagation.

[22]  K. Siakavara,et al.  Application of a Composite Differential Evolution Algorithm in Optimal Neural Network Design for Propagation Path-Loss Prediction in Mobile Communication Systems , 2013, IEEE Antennas and Wireless Propagation Letters.

[23]  Paolo Rocca,et al.  Unconventional Phased Array Architectures and Design Methodologies—A Review , 2016, Proceedings of the IEEE.

[24]  Alex S. Fukunaga,et al.  Improving the search performance of SHADE using linear population size reduction , 2014, 2014 IEEE Congress on Evolutionary Computation (CEC).

[25]  Rainer Storn,et al.  Differential Evolution – A Simple and Efficient Heuristic for global Optimization over Continuous Spaces , 1997, J. Glob. Optim..