Mickaël Pencolé, Eddy Taillefer, Tomoyuki Aono and Takashi Ohira
Direction-of-Arrival Estimation of Two Coherent Signals Employing the ESPRIT Algorithm for ESPAR Antenna: Simulation and Performance Verification
Abstract:The Electronically Steerable Parasitic Array Radiator (ESPAR) antenna is a single output adaptive antenna that offers low cost and low complexity relative to conventional array antennas. A method to estimate direction of arrival (DoA) based on the ESPRIT algorithm has been proposed for a particular 7-element regular hexagonal shaped ESPAR antenna. Nevertheless, this algorithm has never been applied with coherent signals. Therefore, simulations are conducted to investigate the performance of the ESPRIT algorithm for ESPAR antenna in the case of coherent waves. Results show that the DoA estimation performance decreases as the correlation coefficient increases.
In this report, a method to apply the forward-backward averaging (FBA) technique with the ESPRIT algorithm for ESPAR antenna is proposed to improve performance in the case of coherent signals. Simulations show that, in a noise-free environment, the FBA technique can perfectly resolve coherent signals. In a low-noise environment, the proposed technique can improve the acracy of the ESPAR antenna of about 5゜. In the noisy case, the FBA technique does not give a worse performance than the original ESPRIT algorithm for ESPAR antenna. Moreover, the FBA technique provides a criterion for the reliability of the estimates based on the unit-circle. Finally, using the new technique, the computational complexity of the algorithm is not increased. Thus, the proposed method is an improved version of the ESPRIT algorithm for ESPAR antenna.