Long range parametric channel prediction for narrowband MIMO systems with joint parameter estimation

In this paper, we propose a novel long range prediction scheme for narrowband MIMO systems using realistic spatial channel model. The algorithm exploits both the temporal and spatial structure of the MIMO channel to jointly estimate the multipath parameters via a subspace based three dimensional ESPRIT approach. We propose simple transformations to convert the channel impulse response into a space-time manifold matrix exhibiting translational invariance in three dimensions. Simulation results show that the proposed prediction scheme outperforms existing algorithms and can achieve prediction range of several wavelengths.