Estimation of Noncoherently Distributed Channel Parameters of Massive Mimo Systems

This study presents a low complexity signal processing scheme to facilitate the massive multiple input multiple output (massive MIMO) systems to estimate the channel parameters of incoherent distributed channels (including nominal directional of arrival (NDOA) and angular spread (AS)), suppress the multiple access interference and detect multiuse data. Conventional two-dimensional signal processing schemes stack the signal matrices receiving from a two-dimensional antenna array to form high dimensional vectors, and then these vectors are used for succeeding channel parameter estimation and interference suppression, increasing the algorithm’s computation burden. Instead of the way of vector stacking, this study exploits the column and row vectors of the receive signal matrices at the base station, in conjunction with 1D-ESPRIT algorithm and 1D-beamforming to mitigate the computational complexity of the algorithm. In addition, without any pairing processes, the proposed approach can automatically pair the azimuth and elevation estimates, making more suitable for real time applications. On the other hand, the proposed algorithm estimate the AS of each signal cluster through the eigen-values of the signals produced by equalizers based on least square criterion.

Project ID:PB10907-2877
External Project ID:MOST109-2221-E182-042