posted on 2017-12-06, 00:00authored byW Su, WX Zheng, Li Chai
In this paper, blind identification problem based on second-order statistics (SOS) of received signals is investigated for multiple-input multiple-output (MIMO) channels. It is shown that precoders introduced in transmitters can provide certain extra information for blind identification problem and commonly used identifiability conditions on MIMO channels can be relaxed. A design criterion for the precoders in frequency domain is proposed. Two blind identification algorithms based on the knowledge of the precoders and SOS of the received signals are then developed. In these algorithms, due to particular structure of the precoders, the blind identification problem for a MIMO channel is decomposed into several independent single-input multiple-output (SIMO) channel identification problems. The presented algorithms are robust to an overestimated channel order and not sensitive to signal-to-noise ratio (SNR). Compared with the existing results, a larger class of MIMO finite-impulse-response (FIR) channels can be identified by the new methods.
Funding
Category 2 - Other Public Sector Grants Category
History
Volume
54
Issue
7
Start Page
2725
End Page
2736
Number of Pages
12
ISSN
1053-587X
Location
Piscataway, United States
Publisher
Institute of Electrical and Electronics Engineers Inc.