Improvement of the linear blind channel identification algorithms based on second-order statistics
Yiling Xu, Qicong Peng, Yubai Li · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2004
The main blind channel identification methods based on second order statistics include linear prediction algorithm (LPA), outer product decomposition algorithm (OPDA), multi-step linear prediction algorithm (MSLP), least square smoothing algorithm (LSS), and constrained minimum output energy algorithm (CMOE). But the simulations show that the channel couldn't be identified. This paper will analyze these algorithms and point out the reason. The problem happens at the second step. The method based on second order statistics induces an unidentified sign and an indeterminable scale that result in the failure of channel identification. This problem can be resolved in the digital communication system.