Complexity-Reduced Channel Matrix Inversion for MIMO Systems in Time-Varying Channels

Inversion of channel matrix is required for commonly used detection schemes in multiple-input multiple-output (MIMO) systems. In time-varying fading channels, frequent matrix inversion is computationally intensive for mobile terminals operating at high data rates. Several existing papers have addressed this problem for MIMO orthogonal frequency division multiplexing systems by employing interpolation since the channel coefficients are correlated in the frequency domain. The correlation of the channel in the time domain, i.e., the channel matrices at consecutive symbol intervals vary only slightly, could also be exploited. We propose an algorithm that exploits second-order extrapolation in the time domain to lower the computational complexity of matrix inversion. While existing schemes are mainly designed for linear MIMO detection, the proposed algorithm can be applied for both non-linear detection such as ordered successive interference cancelation (OSIC) and linear detection. The proposed scheme can be efficiently implemented with only addition and integer multiplication. Simulation results of the proposed scheme applied in MIMO OSIC detection demonstrate that it can significantly reduce the matrix inversion complexity while maintaining the system performance.