IEEE International Conference on Communications, - Spanning the Universe.
DOI: 10.1109/icc.1988.13542
|View full text |Cite
|
Sign up to set email alerts
|

Spectrally constrained codes for high rate ISDN subscriber loop

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Publication Types

Select...
6
1

Relationship

1
6

Authors

Journals

citations
Cited by 13 publications
(3 citation statements)
references
References 8 publications
0
3
0
Order By: Relevance
“…Interestingly, transmission at two bits per dimension (2 b per baseband symbol or 4 b per QAM symbol) has been found to yield comparable or better results than other considered rates in a number of different HDSL transmission studies [12], [70]- [73]. To date, there are several manufacturers offering HDSL equipment employing different transmission technologies.…”
Section: B Hdslmentioning
confidence: 99%
“…Interestingly, transmission at two bits per dimension (2 b per baseband symbol or 4 b per QAM symbol) has been found to yield comparable or better results than other considered rates in a number of different HDSL transmission studies [12], [70]- [73]. To date, there are several manufacturers offering HDSL equipment employing different transmission technologies.…”
Section: B Hdslmentioning
confidence: 99%
“…We require several thousand coefficient locations; each of which, fortunately, need only be accessed every NT time units, which actually leaves the BDFE slightly more computationally efficient than a conventional DFE [lo]. In [9] we presented some basic ideas on channel modeling. We identified the fact that the channel could be accurately modeled with a small number of parameters by using a pole-zero model instead of an all-zero model.…”
Section: Pre-/post-equalizationmentioning
confidence: 99%
“…Finally, we could avoid the zersforcing approach altogether and allow subsequent blocks to interfere while minimizing the power of the error signal. In [9] we assumed that the prefilter was designed to cancel the poles and zeros of the channel for a subset of possible input vectors. The problem with this assumption is that knowledge of the channel is required at the input both to form the prefilter and to determine the optimum signaling vectors.…”
Section: Vector Codingmentioning
confidence: 99%