A New DOA Algorithm for Spectral Estimation

Abstract

Uniform linear arrays (ULAs) are the most commonly used configurations for the direction-of-arrival estimation of radio signals in many wireless communication applications due to its simplicity. However, this configuration does not provide uniform estimation and deteriorates in performance for the grazing incidences. In this work, we solve this problem by modifying the structure of a conventional ULA. This new confirmation has two extra antenna elements at the ends of the array. These extra elements will automatically activate when the source signals are impinging for the grazing incidences. We modify the multiple signal classical algorithm and develop an improved array signal model to obtain the high-resolution pseudo-spectrum of the received signals for full azimuth angles. The performance of the proposed method is compared with the various well-known methods in many practical scenarios. Finally, total Root Mean Square Error is examined over a wide range of Signal to Noise Ratio. Simulation results show that the proposed method outperforms the well-knows methods and provides the accurate and uniform estimation of narrowband signals for end fire array angles which fall in the range $\left( { - 70^{{^\circ }} \le \theta \le - 90^{{^\circ }} } \right)$ and $\left( {70^{{^\circ }} \le \theta \le 90^{{^\circ }} } \right)$, at middle angles $\left( { - 60^{{^\circ }} \le \theta \le 60^{{^\circ }} } \right)$ and at boresites which is very useful in 5G and beyond mobile communication.