Space-time structure of diffused multipath channels

In wireless communication systems with specular multipath channels, the channel structure can be used to improve the system performance.] However, the channel structure turns obscure for diffused multipath channels. We analyze the diffused multipath channels and develop a low-complexity algorithm to determine the optimal dimension of the approximated channel subspace as well as the optimal basis for the subspace, in the sense of least-square error of the channel estimates. Simulation results show that (1) the system performance is significantly improved by exploring the structure of the diffused multipath channels, (2) the performance improvement is even more significant in a severe environment where the signal to noise ratio is low.