Research On Parameter Estimation Of MIMO Channel In The Presence Of Mutual Coupling And Coherent Signals

The MIMO communication system has attracted growing interest in recent years for its high speed and reliability. To get aware of the channel's characters and ensure the transmission quality of wireless signals, we need to obtain accurate estimations of channel parameters including direction, delay etc with low complexity. However, performances of conventional algorithms degrade in the presence of mutual coupling, signal coherency and colored Gaussian noise. Therefore it is of great importance to study methods which can suppress mutual coupling, coherency and colored Gaussian noise simultaneously.In this paper, a novel direction of arrival(DOA) estimation approach is proposed for coherent signals against array mutual coupling in colored Gaussian noise situation. Firstly, auxiliary sensors are adapted to suppress mutual coupling effect. Then the spatial smoothing technique and fourth-order cumulants are combined to eliminate the Gaussian noise and simultaneously solve signal coherency. Finally, in the estimation process of DOAs, the ESPRIT method is adopted. Simulation results demonstrate that the proposed algorithm can accurately estimate DOAs in arbitrary Gaussian noise situation and that its performance can be very close to the case without any coupling.Further, to avoid the computational complexity donated by the constructed sub-matrices of spatial smoothing, a new algorithm constructing a Toeplitz matrix, which is independent of the signal coherency, is presented. Complexity analysis proves this approach has lower computational cost than the former and can provide precise DOA estimations. Besides, the estimation performance of this algorithm is close to that of the optimum case without coupling. Comparing with the method employing spatial smoothing, this algorithm has a better estimation performance.In summary, this paper studies the parameter estimation methods for MIMO channel in the presence of mutual coupling, signal coherency and arbitrary Gaussian noise. In addition to theoretical analysis, the effectiveness of the proposed algorithms is verified by simulations.