Proceedings of the Tenth IEEE Workshop on Statistical Signal and Array Processing (Cat. No.00TH8496)
DOI: 10.1109/ssap.2000.870205
|View full text |Cite
|
Sign up to set email alerts
|

Global convergence of a single-axis constant modulus algorithm

Abstract: We propose a modification to the Constant Modulus Criterion for real valued sources processed with a complex valued receiver. Our modification is called Single-Axis Constant Modulus Criterion (SA-CM) because it operates solely on the real component of the complex equalizer output. We show that under idealized conditions, a finite length, baudspaced, complex valued equalizer minimizing the SA-CM criterion admits only desirable global minima settings that are ISI-free. A single-axis receiver architecture is comp… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Publication Types

Select...
3
1

Relationship

0
4

Authors

Journals

citations
Cited by 4 publications
(1 citation statement)
references
References 9 publications
0
1
0
Order By: Relevance
“…(15)] for the case of an infinite-length CM equalizer, under the simplifying assumption of BPSK symbols with unitary variance. Finally, it is noteworthy that with reference to real-valued symbols (i.e., ), it was shown in [12] that the undesired minima (8) and (9) disappear by minimizing the CM cost function (2), provided that the equalizer output is not a strictly linear function of ; namely, . More generally, if the transmitted symbols are improper complex, one has to use widely linear equalizing structures [5], where the equalizer output is given by and the CM cost function is minimized with respect to both and , where is not necessarily constrained to be equal to .…”
Section: Analysis Of the CM Cost Functionmentioning
confidence: 99%
“…(15)] for the case of an infinite-length CM equalizer, under the simplifying assumption of BPSK symbols with unitary variance. Finally, it is noteworthy that with reference to real-valued symbols (i.e., ), it was shown in [12] that the undesired minima (8) and (9) disappear by minimizing the CM cost function (2), provided that the equalizer output is not a strictly linear function of ; namely, . More generally, if the transmitted symbols are improper complex, one has to use widely linear equalizing structures [5], where the equalizer output is given by and the CM cost function is minimized with respect to both and , where is not necessarily constrained to be equal to .…”
Section: Analysis Of the CM Cost Functionmentioning
confidence: 99%