Title :
A system-theoretic foundation for blind equalization of an FIR MIMO channel system
Author :
Inouye, Yujiro ; Liu, Ruey-wen
Author_Institution :
Dept. of Electron. & Control Syst. Eng., Shimane Univ., Matsue, Japan
fDate :
4/1/2002 12:00:00 AM
Abstract :
The system-theoretic foundation for blind equalization of a finite-impulse response (FIR) MIMO channel system in the channel-filter setting is investigated. In this setting, a general definition of equalizable channel system is presented, and its characterization is given: Every equalizable channel system has an FIR irreducible-paraunitary factorization. Based on this fact, we show that any equalizable channel system can be reduced to a paraunitary FIR system by a filter, called a whitener. It is shown that one such whitener can be designed by a linear prediction-based approach. It is also shown that if the original FIR channel system has no transmission zero at the origin, then the above paraunitary FIR channel system is a static system
Keywords :
FIR filters; Hilbert spaces; MIMO systems; blind equalisers; deconvolution; filtering theory; matrix decomposition; parameter estimation; polynomial matrices; prediction theory; transfer function matrices; FIR filters; Hilbert space; blind equalization; blind signal separation; blockers; canonical structures; channel-filter setting; deconvolvers; demixers; equalizable channel system definition; finite-impulse response MIMO channel system; irreducible-paraunitary factorization; linear prediction-based approach; polynomial matrix; second-order statistics; system-theoretic foundation; transfer function; transmission zeros; whitener; Antenna arrays; Antennas and propagation; Blind equalizers; Blind source separation; Delay; Filtering theory; Finite impulse response filter; MIMO; Statistics; Transmitting antennas;
Journal_Title :
Circuits and Systems I: Fundamental Theory and Applications, IEEE Transactions on
