Robust blind channel estimation for Multiple-Input Multiple-Output Orthogonal Frequency-Division Multiplexing
Blind channel estimation for MIMO-OFDM systems based on blind source separation (BSS) is currently an active area of research. A blind channel estimator for MIMO-OFDM systems using second-order blind identification (SOBI) algorithm was previously presented in the literature. The SOBI algorithm relie...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2013
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Blind channel estimation for MIMO-OFDM systems based on blind source separation (BSS) is currently an active area of research. A blind channel estimator for MIMO-OFDM systems using second-order blind identification (SOBI) algorithm was previously presented in the literature. The SOBI algorithm relies only on the second-order statistics (e.g., correlation) of the received signal. It is well known that second-order statistics is not sufficient for BSS. Recently, a BSS algorithm based on correntropy independent components analysis (ICA) was formulated for separating both independent and identically distributed (i.i.d) sources and temporally correlated sources with distinct spectra. Correntropy is a generalised correlation function between two random processes. Unlike the conventional correlation function, the correntropy function contains both the higher-order statistics and temporal structure of the random processes. Motivated by the property of the correntropy, we propose a blind channel estimator based on correntropy ICA for MIMO-OFDM and show that its performance is better than that of SOBI based method. To the best of our knowledge, this is the first work which applies correntropy ICA in blind MIMOOFDM channel estimation. |
|---|