Logarithmic-domain array interpolation for improved DOA estimation in automotive radars

Abstract

In this paper, we propose a novel array interpolation method in logarithmic domain for enhanced direction of arrival (DOA) estimation. Array interpolation techniques are broadly utilized in the DOA estimation for achieving better angular resolution. Generally, to generate interpolated array elements from original array elements, the method of linear least squares (LLS) has been used. When we use the LLS method, the interpolated array elements are formulated by linear combinations of the original array elements. Therefore, amplitudes of the interpolated array elements may not be equivalent to those of the original array elements. In addition, through a transformation matrix from the LLS method, phases of the interpolated array elements are not precisely generated. For the DOA estimation, phase information of interpolated array elements is important. Therefore, we propose a novel array transformation matrix for generating accurate phases of the interpolated array elements to improve the DOA estimation performance. Verifying from simulation and measurement results, our method shows much better angular resolution and estimation accuracy.