Adaptive Millimeter-Wave Channel Estimation and Tracking

Computationally efficient channel estimation is critical to optimize the capacity of millimeter-wave communication network. For this, compressed-sensing has been recommended to estimate a few dominant channel parameters. However, it is challenging to extend compressed-sensing solutions to a continuous channel tracking due to the number of required measurements. In this paper, we propose efficient adaptive channel estimation and tracking for millimeter-wave communication with minimum communication overhead. We recommend characterizing the instantaneous rate of change of the millimeter-wave channel as a gradient of spectral overlap between channels. The significant channel variations are then detected locally by applying convergence in mean square sense to the resultant time sequence. If the channel experiences significant variation, then the multipath components are estimated directly using compressed-sensing. Otherwise, the channel parameters are updated using the channel tracking model. For this purpose, we introduce an efficient channel tracking model based on small-angle assumption. The proposed channel tracking method employs an autoregressive process to update the angle of departure and angle of arrival. We present numerical results to evaluate the proposed adaptive channel estimation and tracking method.