Deep Learning Approach for Optimal Angle-of-Arrival Estimation Using a mm-Wave Sensor

From automation industries to health care setups, indoor localization is becoming a key necessity to enable the fourth industrial revolution. With its rising demand, the need for more precise positioning systems is increasing day by day. Millimeter-wave (mm-Wave) technology is emerging to enable highly precise localization performance. However, due to the limited availability of low-cost mm-Wave sensors, it is challenging to accelerate research on real data. Furthermore, noise due to the hardware components of a sensor incurs perturbation in the received signal, which corrupts the estimation of the angle of arrival (AoA). Therefore, we propose a data-driven approach, which employs a dense neural network (DNN) to reduce errors in the estimate of AoA obtained when using a mm-Wave sensor. Our main goal is to optimize the measurements acquired from low-cost mm-Wave sensors to accelerate the development of proof of concepts and foster research on mm-Wave based indoor positioning systems. Our experimental results show a 50% decrease in the error of AoA estimation using the proposed DNN-based approach. All experiments have been performed on over-the-air data collected using a mm-Wave sensor.