A comparative study among swarming intelligence algorithms and subspace based algorithms for high resolution direction of arrival estimation

Recent advances in underwater signal processing have enabled the parameter estimation of Direction of Arrival (DOA) utilizing evolutionary computing paradigms, which also has witnessed several applications in the field of seismology, earthquakes, astronomy, and biomedicine. In this work, an innovative study of comparison among state of the art subspace-based and particle swarm optimization (PSO) and Genetic Algorithms (GA) algorithms is presented to have effective DOA estimates for various objects with dynamical characteristics in underwater scenario. The viability of innovative statistical indices is employed to describe performance in order to evaluate it. The effectiveness of the GA and PSO a is evaluated in comparison with its traditional counterparts (such as MVDR, MVDR, MUSIC, ESPRIT and UESPRIT) based on different metrics such as estimation accuracy, probability of resolution and computational robustness against the number of elements and noise. For validation assessments, Crammer Rao Bound (CRB) analysis is also performed, and outcomes from Monte Carlo runs show that the Genetic Algorithm (GA) outperforms its analogue in terms of complexity indices, convergence, and precision.