Spatial Zadoff-Chu Modulation for Rapid Beam Alignment in mmWave Phased Arrays

摘要

Compressed sensing (CS) algorithms can be used to perform beam alignment in mmWave phased arrays with fewer channel measurements by exploiting the sparsity of mmWave channels. Unfortunately, several efficient CS matrices cannot be realized in phased arrays due to hardware constraints. In this work, we propose to use shifted Zadoff-Chu (ZC) sequences in the antenna domain to realize efficient CS matrices for channel estimation or beam alignment. We prove that the shifted ZC-based CS matrix satisfies the restricted isometry property with high probability. Furthermore, CS algorithms that use the proposed training can exploit the fast Fourier transform and are computationally efficient over those that use pseudo-random phase shifts. Using simulations, we show that the proposed CS matrix achieves better beam alignment performance than the commonly used random phased shift-based CS matrices.