ANN Model for DoA Estimation of the Signal Received by Crumpled Textile Wearable Antenna Array

The ANN-based architecture of the module for fast DoA estimation of the signal received by two-element textile wearable antenna array (TWAA) is proposed in this paper. Developed MLP DoA module includes the effect of changing the distance between the antenna elements during the movement of the TWAA wearer and the textile crumpling. Elements of the spatial correlation matrix of the signal sampled by TWAA are fed to the input of the DoA module. The output of the DoA module is the angular position of the signal source (radio gateway) in the azimuthal plane relative to the TWAA. After the appropriate training of the MLP network and implementation of the MLP DoA module, the performance of the module under the conditions of changing the distance between the antenna elements and changing the signal-to-noise ratio is investigated.