A statistical approach to subspace based blind identification
1998 (English)In: IEEE Transactions on Signal Processing, ISSN 1053-587X, E-ISSN 1941-0476, Vol. 46, no 6, 1612-1623 p.Article in journal (Refereed) Published
Blind identification of single input multiple output systems is considered herein. The low-rank structure of the output signal is exploited to blindly identify the channel using a subspace fitting framework. Two approaches based on a minimal linear parameterization of a subspace are presented and analyzed. The asymptotically best consistent estimate is derived for the class of blind subspace-based techniques. The asymptotic estimation error covariance of the subspace estimates is derived, and the corresponding covariance of the statistically optimal estimates provides a lower bound on the estimation error covariance of subspace methods. A two-step procedure involving only linear systems of equations is presented that asymptotically achieves the bound. Simulations and numerical examples are provided to compare the two approaches.
Place, publisher, year, edition, pages
1998. Vol. 46, no 6, 1612-1623 p.
Blind idenification, equalization, single-input multiple output system, statistical analysis, subspace fitting
IdentifiersURN: urn:nbn:se:kth:diva-60026DOI: 10.1109/78.678476ISI: 000073770000012OAI: oai:DiVA.org:kth-60026DiVA: diva2:477143
NR 201408052012-01-122012-01-122012-02-14Bibliographically approved