Nonlinear complex M-SVR for LTE MIMO-OFDM channel with impulsive noise

In this paper, an efficient approach is proposed to track the double-selected multipath channel for a Multiple-Input-Multiple-Output Orthogonal-Frequency-Division-Multiplexing (MIMO-OFDM) system in the existence of both Gaussian and non-Gaussian noises interfering with reference signals. Using nonlinear complex Multiple Support Vector Machine Regression (M-SVR) methodology, fast-fading multipath channel estimation can be performed accurately. The M-SVR algorithm is developed and affected to MIMO Long Term Evolution (LTE) downlink system with Space Time Block Coding (STBC) scheme. The effectiveness of the proposed technique is demonstrated through the obtained simulation results which confirm the abilities of the M-SVR to track the fast-fading channel in the manifestation of several nonlinearities.