Artificial Neural Network Based Classification Model For Minimum Redundancy Array and Minimum Hole Array

The Neural Networks(NN) is a subset of Machine Learning (ML) Algorithms used for prediction and function approximation. In this paper, we propose a Artificial Neural Network(ANN) based architecture for fast Direction of Arrival(DOA) estimation of signal received by by both Minimum Redundancy Array (MRA) and Minimum Hole Array (MHA) that is used to estimate bearing angle (or) angle of arrival of the receieved signals. The integration of concepts of Sparse arrays and Deep Neural Network are proposed in this paper. The number of sensors for both arrays MRA and MHA used in this simulation are $\left| \mathbb{S} \right| = 6$ Sensors respectively. After the appropriate training of the ANN architecture, the performance of model is evaluated under a set of parameters mentioned in table for both MRA and MHA, the precison, recall, f1 score are evaluation metrics with which the model is evaluated, the model performed slightly better with MRA as input than MHA as input.