PMID- 31510445
OWN - NLM
STAT- PubMed-not-MEDLINE
DCOM- 20190912
LR  - 20190912
IS  - 1094-4087 (Electronic)
IS  - 1094-4087 (Linking)
VI  - 27
IP  - 18
DP  - 2019 Sep 2
TI  - Machine learning classifier based on FE-KNN enabled high-capacity PAM-4 and NRZ 
      transmission with 10-G class optics.
PG  - 25802-25813
LID - 10.1364/OE.27.025802 [doi]
AB  - Modified by special feature engineering, a powerful and low-order equalizer based 
      on K-nearest neighbors (KNN) classifier is applied to improve performance of 
      high-speed system with bandwidth-limited optics. The feature construction and 
      feature weighting are specially designed to conduct an appropriate a feature 
      engineering-based KNN (FE-KNN) scheme, which contains more data characteristics 
      to enhance the equalization performance. Experimental comparisons of KNN 
      classifier with/without feature engineering, decision feedback equalizer (DFE) 
      and feed-forward equalizer (FFE) are implemented to prove the feasibility of our 
      scheme in both 25-Gb/s NRZ and 50-Gb/s PAM-4 transmission experiments with 10-G 
      optics system. The corresponding results show that, without the feature 
      engineering, the performance achieved by the common KNN is not improved even in 
      the case of hard decision (HD). In contrast, compared to the common 11-taps DFE, 
      the performance achieved by FE-KNN with only 5 taps is improved by 1-dB at 
      KP4-FEC threshold (BER=2.2E-4) for 25-Gb/s NRZ transmission. While, for 50-Gb/s 
      PAM-4 case, 0.5-dB sensitivity improvement is achieved by our scheme compared to 
      the common 11-taps DFE under the HD-FEC limit (BER=3.8E-3).
FAU - Bi, Meihua
AU  - Bi M
FAU - Yu, Jiasheng
AU  - Yu J
FAU - Miao, Xin
AU  - Miao X
FAU - Li, Longsheng
AU  - Li L
FAU - Hu, Weisheng
AU  - Hu W
LA  - eng
PT  - Journal Article
PL  - United States
TA  - Opt Express
JT  - Optics express
JID - 101137103
SB  - IM
EDAT- 2019/09/13 06:00
MHDA- 2019/09/13 06:01
CRDT- 2019/09/13 06:00
PHST- 2019/09/13 06:00 [entrez]
PHST- 2019/09/13 06:00 [pubmed]
PHST- 2019/09/13 06:01 [medline]
AID - 417187 [pii]
AID - 10.1364/OE.27.025802 [doi]
PST - ppublish
SO  - Opt Express. 2019 Sep 2;27(18):25802-25813. doi: 10.1364/OE.27.025802.