Fast DoA estimation of the signal received by textile wearable antenna array based on ANN model

Abstract

MLP_DoA module, being an integral part of the smart TWAA DoA subsystem, intended for fast DoA estimation is proposed. Multilayer perceptron network is used to create the MLP_DoA module that provides a radio gateway location in azimuthal plane at its output when a spatial correlation matrix, found by receiving the radio gateway signal using two-element textile wearable antenna array, is on its input. MLP_DoA network training with monitoring the generalization capabilities on the validation set of samples is applied. The accuracy of the proposed modeling approach is compared to the classical approach in MLP_DoA module training previously developed by the authors. Comparison of the presented ANN model with the root MUSIC algorithm in terms of accuracy and program execution time is also done.

Advanced ANN model for DoA estimation in smart textile wearable antenna array subsystem

“…The normalization of the matrix C does not change the results of the MUSIC algorithm for DoA estimation. 26,29 By introducing the variable g i -root gain ratio of the ith antenna element, as…”

“…The spatial correlation matrix is estimated from a large number of TWAA output samples in a short time interval (TWAA snapshots) using fast A/D converters and the matrix elements are calculated on the FPGA module using the following approximate formula. 29 C ≈ 1 N ss…”

“…The following metrics were used in the neural network testing: worst case error (WCE), average test error (ATE), Pearson product moment correlation coefficient (r PPM ), and root mean square error (RMSE). 23,29 The worst case error is calculated as follows…”

“…MLP networks with two hidden layers (H = 2, MLP2-N 1 -N 2 ) were considered for the realization of the DoA modules, since good results in DoA estimation have already been obtained with their application . [28][29][30] In order to find an MLP network architecture for each case that performs the DoA estimation as accurately as possible, a number of different MLP2-N 1 -N 2 architectures were trained, where N 1 and N 2 . 8,22 After the training process, each trained network was tested with the test set, and based on the test metrics given in the previous section, a specific MLP network was selected for each case to realize the ANN-based DoA module (MLP DoA module).…”

“…[28][29][30] In these papers, the multilayer perceptron (MLP) network was used to create these basic versions of the so-called MLP DoA module, but with some simplifications that do not fully correspond to the real-case scenarios. For example, in, 28,29 this module was developed assuming that only the gain of each antenna element changes due to the textile creasing, while neglecting the other physical changes in the TWAA, such as the change of the distance between the antenna elements, which significantly affect the DoA estimation accuracy. On the other hand, in the version of the MLP DoA module developed in, 30 only the change in the distance between two antenna elements was considered and the change in the gains of the antenna elements was neglected.…”