State-Space Modeling and Propagation Parameter Tracking: Multitarget tracking based approach

The paper describes a state-space approach for retrieving the parameters of the double directional MIMO propagation channel model from channel sounding measurements. We address the issues arising from tracking a varying number of jointly estimated targets (propagation paths) from a vast amount of data. We focus on state dimensionality estimation, i.e., how to drop paths from the state as well as augmenting the state with new path estimates. We propose a whiteness test for detecting the time instances when to increase the number of paths to track. Simulation results are presented to illustrate the benefits of the path detection algorithm.