Learning from data: Applications of Machine Learning in optical network design and modeling

Hernández, José Alberto

doi:10.23919/ondm48393.2020.9133007

Cited by 4 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, our study has only 92 sample points; in the study of Aptoula, E. et al, the number of sample points reached 320 and their R 2 was higher than ours [57]. Although we increased the sample points via data expansion, more measured sample points could lead to higher accuracy [70]. Limited by the current sensor development, there are fewer satellites with both high spatial resolution and good spectral characteristics [71].…”

Section: Limitationsmentioning

confidence: 84%

Water Quality Chl-a Inversion Based on Spatio-Temporal Fusion and Convolutional Neural Network

Yang

Zhao

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

The combination of remote sensing technology and traditional field sampling provides a convenient way to monitor inland water. However, limited by the resolution of remote sensing images and cloud contamination, the current water quality inversion products do not provide both high temporal resolution and high spatial resolution. By using the spatio-temporal fusion (STF) method, high spatial resolution and temporal fusion images were generated with Landsat, Sentinel-2, and GaoFen-2 data. Then, a Chl-a inversion model was designed based on a convolutional neural network (CNN) with the structure of 4-(136-236-340)-1-1. Finally, the results of the Chl-a concentrations were corrected using a pixel correction algorithm. The images generated from STF can maintain the spectral characteristics of the low-resolution images with the R2 between 0.7 and 0.9. The Chl-a inversion results based on the spatio-temporal fused images and CNN were verified with measured data (R2 = 0.803), and then the results were improved (R2 = 0.879) after further combining them with the pixel correction algorithm. The correlation R2 between the Chl-a results of GF2-like and Sentinel-2 were both greater than 0.8. The differences in the spatial distribution of Chl-a concentrations in the BYD lake gradually increased from July to August. Remote sensing water quality inversion based on STF and CNN can effectively achieve high frequency in time and fine resolution in space, which provide a stronger scientific basis for rapid diagnosis of eutrophication in inland lakes.

show abstract

Section: Limitationsmentioning

confidence: 84%

Water Quality Chl-a Inversion Based on Spatio-Temporal Fusion and Convolutional Neural Network

Yang

Zhao

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Namely, EF class is preferred in comparison with AF subclasses and BE traffic class, while AF subclasses are prioritized over BE traffic class. Among AF subclasses AF4 has the highest priority followed by AF3 subclass and then AF2 and AF1 respectively ( 20)- (25). In the last step the excess bandwidth calculated for each class is allocated to ONUs based on previously calculated weight factors (26).…”

Section: T Cyclementioning

confidence: 99%

“…Unlike single-channel systems, QoS support in WDM EPONs was considered in a significantly smaller number of proposed solutions [18][19][20] although various dynamic wavelength bandwidth allocation (DWBA) algorithms have been proposed to date. More recently, artificial intelligence (AI) and machine learning (ML) techniques and algorithms have gained significant attention, promising to reduce the complexity of network and the requirement of manual operations [21][22][23][24][25]. However, these models are mainly focused on achieving better utilization and do not discuss different traffic types/profiles and their particular needs.…”

Section: Service Evolutionbackground Und Motivationmentioning

confidence: 99%

The way forward truly QoS aware EPON

Radivojević

Matavulj

2022

Preprint

View full text Add to dashboard Cite

In the last decade, the bit rates achieved on fiber broadband networks have increased multiple times and every new generation of passive optical network (PON) technologies must unlock more of fiber’s potential and must be backward compatible to ensure easy and iterative upgrade of operator’s networks. In this paper we present two novel dynamic wavelength (in the case of multichannel system) and bandwidth allocation models for realization of fully QoS (Quality of Service)-aware single-channel i.e. multichannel Ethernet passive optical network (EPON). The both presented models include a novel approach for inter/intra –ONU (Optical Network Unit) scheduling mechanism as well as advanced support for multimedia and business-related services. Moreover, the single channel system could be upgraded to multichannel system on per-need basis i.e., only nodes with higher traffic demands could be upgraded. The aim of this approach is twofold. On the one side we hope to avoid the implementation of additional, complex algorithms usually related to QoS support in the access segment and on the other side to develop a solution in which new and old EPON technologies are able to coexist. This approach would eliminate the need for complex migrations, complicated and time consuming procedures as well as possible service outages in the access network. We present detailed simulation experiments with the aim of conducting a study of the performances of both presented approaches and validate both of them in the terms of the future application, further service development and implementation in the next generation broadband networks (NGNs).

show abstract

The way toward truly QoS-aware EPON

Radivojević,

Matavulj

2024

Photon Netw Commun

View full text Add to dashboard Cite

Learning from data: Applications of Machine Learning in optical network design and modeling

Cited by 4 publications

References 25 publications

Water Quality Chl-a Inversion Based on Spatio-Temporal Fusion and Convolutional Neural Network

Water Quality Chl-a Inversion Based on Spatio-Temporal Fusion and Convolutional Neural Network

The way forward truly QoS aware EPON

The way toward truly QoS-aware EPON

Contact Info

Product

Resources

About