Empirical modeling of cross-correlation for spatial-polarimetric channels in indoor scenarios

In this paper, propagation channel measurement data obtained by using the wide-band multiple-input-multiple-output (MIMO) channel sounder - PROPsound - is applied to analyzing the cross-correlation behavior of spatial-polarimetric (SP) channels in indoor environments. Four antennas are selected from the transmitter (Tx) array and from the receiver (Rx) array to form a 2×2 spatial MIMO channel, with each spatial element consisting of co-located ±45°-polarized antennas. The characteristics of the resultant 16 × 16 narrow-band SP Channel Correlation Matrix (CCM) are studied for five indoor environments of different types. Results show that the Kronecker model proposed in the 3rd Generation Partnership Project (3GPP) TS36.101, which describes the SP CCM as the Kronecker product of the separate Tx and Rx spatial CCMs, and the 4×4 polarimetric CCM, is inapplicable in indoor scenarios. As an alternative, we propose to consider the cross-correlation coefficient of SP channels as a random variable, and apply its empirical statistics for stochastic modeling. Results also show that the empirical distribution of the off-diagonal elements of SP CCM is distinctive with respect to the types of environments.