DocumentCode :
1722726
Title :
Channel gain change detection and channel profile selection in a multicarrier system
Author :
Salvekar, Atul ; Aldana, Carlos ; Tellado, Jose ; Cioffi, John
Author_Institution :
STAR Lab., Stanford Univ., CA, USA
Volume :
2
fYear :
1999
fDate :
6/21/1905 12:00:00 AM
Firstpage :
1133
Abstract :
This paper investigates the detection and identification of abrupt channel gain changes in multicarrier systems, which can be particularly noticeable in the area known as splitterless digital subscriber lines. Gross channel changes nominally lead to a need for retraining of the modems in the connection, but this paper introduces some mechanisms known as profile detection that can avert the need for retraining in a number of realistic situations. Specific methods are introduced that allow for the detection of an abrupt gain change and further minimize the probability of error in selecting from one of a set of profiles that have been previously observed (or conjectured) to characterize the possible modes of operation of the channel. Expressions for analysis are also provided
Keywords :
digital subscriber lines; modulation; parameter estimation; signal detection; telecommunication channels; ADSL; AWGN; MAP detection; asymmetric digital subscriber loops; blind estimator; channel gain change detection; channel gain change identification; channel profile selection; discrete multitone; error probability; gross channel changes; maximum a posteriori detection; modem retraining; multicarrier modulation; multicarrier system; profile detection; splitterless digital subscriber lines; Channel estimation; DSL; Delay; Event detection; Intersymbol interference; Modems; OFDM modulation; Quadrature amplitude modulation; Telecommunications; Time frequency analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Global Telecommunications Conference, 1999. GLOBECOM '99
Conference_Location :
Rio de Janeireo
Print_ISBN :
0-7803-5796-5
Type :
conf
DOI :
10.1109/GLOCOM.1999.829950
Filename :
829950
Link To Document :
بازگشت