2008
DOI: 10.1109/jlt.2008.920135
|View full text |Cite
|
Sign up to set email alerts
|

Performance Analysis of MLSE Receivers Based on the Square-Root Metric

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
9
0
18

Year Published

2009
2009
2025
2025

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 23 publications
(27 citation statements)
references
References 27 publications
0
9
0
18
Order By: Relevance
“…In various MLSE systems, both through simulations and experiments [34]- [36], we found that the following law fitted well the needed processor memory vs. CD, for smooth NRZ pulses:…”
Section: Number Of Fir Filter Taps For Pm-qammentioning
confidence: 73%
“…In various MLSE systems, both through simulations and experiments [34]- [36], we found that the following law fitted well the needed processor memory vs. CD, for smooth NRZ pulses:…”
Section: Number Of Fir Filter Taps For Pm-qammentioning
confidence: 73%
“…In the latter case, memory is introduced due to the interaction of fiber nonlinearities, dispersion and noise, which cannot be completely removed by the DSP chain. Detectors for channels with memory are typically based on the maximum-likelihood sequence estimation [43], [44], [45], [46], whose complexity grows exponentially with both the memory and the alphabet cardinality (and thus the target channel use efficiency). This constrains standard, fast optical receivers to memoryless processing, effectively applying the mismatched decoding principle in the following way.…”
Section: Achievable Information Ratesmentioning
confidence: 99%
“…The MLSE runs at 1 sample per symbol and operates statically; meaning it is trained at the beginning and not changed afterwards any more. Along the lines of [25] we estimate the mean of the probability density functions (PDFs) of a channel matrix using 2 15 received samples and the corresponding digital data from the first period of the sent sequence. This assumes an additive Gaussian noise distribution with equal variance for all PDFs.…”
Section: Mlsementioning
confidence: 99%