Two-step lattice super-exponential algorithm for blind equalization of multi-input multi-output channels

Abstract

Feng and Chi (1998) reported a two-step lattice super-exponential algorithm (2S-LSEA) for blind equalization of single-input single-output (SISO) channels that is superior to Shalvi and Weinstein's (1993) FIR filter based super-exponential algorithm (SEA) in faster convergence speed, lower computational complexity, and more reliable performance to a variety of channels, besides modularity and low sensitivity to parameter quantization effects of lattice structure. In this paper, a 2S-LSEA for multi-input multi-output (MIMO) channels is proposed that is also superior to Yeung and Yau's (1998) SEA for MIMO channels in the same preceding advantages of the 2S-LSEA for SISO channels. Some simulation results are presented to support the efficacy of the proposed 2S-LSEA for MIMO channels.