A Joint DOA and TOA Estimation Scheme for 5G Signals Under Array Modeling Errors

Abstract

This paper addresses the problem of direction-of-arrival (DOA) and time-of-arrival (TOA) estimation in the presence of array modeling errors for the fifth generation (5G) signals. A joint estimation scheme based on the two-dimensional multiple signal classification (MUSIC) algorithm is proposed. It employs a pre-calibrated space-frequency manifold rather than the ideal closed-form manifold to counteract the direction-dependent array modeling errors. However, the pre-calibrated manifold does not exhibit the Vandermonde structure in angular domain even for uniform arrays, preventing the usage of the spatial smoothing technique. Since the 5G signal usually occupies a large bandwidth with a wealth of subcarriers, a modified smoothing strategy which only smooths in the frequency-domain is proposed to recover the rank of the covariance matrix. The proposed scheme is demonstrated to calibrate the direction-dependent array errors accurately in a multipath-free anechoic chamber. Its effectiveness in multipath-harsh indoor environments is also validated by both the simulated and the real-measured 5G indoor channel data.