A MIMO turbo equalizer for frequency-selective channels with unknown interference

Tetsushi Abe, Shigeru Tomisato, Tad Matsumoto

Research output: Contribution to journalArticlepeer-review

45 Citations (Scopus)


A new space-time turbo equalization algorithm is derived for frequency-selective multiple-input-multiple-output (MIMO) channels with unknown interference. The algorithm is an extension of our proposed MIMO equalization algorithm [12], which performs joint channel estimation, multiple user's signal detection, and decoding, all in an iterative manner. This paper's proposed algorithm uses estimates of the correlation matrix of composite unknown interference-plus-noise components to suppress the unknown interference while effectively separating multiple users' signals to be detected (referred to as "known user" later). The correlation matrix of the composite unknown interference-plus-noise components can be estimated by time averaging the instantaneous emperical correlation matrix over the training period. Since the iterative channel estimation yields better channel estimates as more iterations are performed, thereby the estimate of the correlation matrix of the unknown interference-plus-noise components also becomes more accurate. This results in better signal detection performances, even in the presence of unknown interferers. A series of computer simulations show that this paper's proposed algorithm can properly separate known users' signals while suppressing unknown interference.

Original languageEnglish
Pages (from-to)476-482
Number of pages7
JournalIEEE Transactions on Vehicular Technology
Issue number3
Publication statusPublished - May 2003
Externally publishedYes


  • Multiple-input-multiple-output (MIMO) system
  • Space-time signal processing
  • Turbo equalization

ASJC Scopus subject areas

  • Automotive Engineering
  • Aerospace Engineering
  • Electrical and Electronic Engineering
  • Applied Mathematics


Dive into the research topics of 'A MIMO turbo equalizer for frequency-selective channels with unknown interference'. Together they form a unique fingerprint.

Cite this