The article shows the results of the algorithm of Convolutional Neural Networks (CNN) based on data from the Asynchronous Delay Tap Sampling (ADTS) method. The combination of these two techniques made it possible to obtain an expert system to monitor simultaneously impairments (Chromatic Dispersion, Crosstalk and Optical to Signal Noise Ratio) occurring during transmission in the physical layer of the optical network. The ADTS technique allows to present disturbances in the form of images called phase portraits. Using the VPIphotonics program, 62000 images for modulation OOK were generated. All images were disturbed by a different combination of CD, Crosstalk and OSNR phenomena. These combination contain interference in pairs (e.g. 25 ps/nm for CD, 24.5 dB for Crosstalk and 17.5 dB for OSNR) but also individually (e.g. 678.93 ps/nm for CD and 0 for Crosstalk and OSNR). The next stage was the use of the convolutional algorithms of neural network. Research was conducted to teach the network the best possible recognition of simultaneously occurring phenomena. The challenge in the field of monitoring optical networks is monitoring with high accuracy simultaneously occurring phenomena using one method. As a result of the conducted research, high possibilities of simultaneous monitoring of the CD and Crosstalk phenomena for which the đť‘… 2 match factor was obtained at the level higher than 0.999 were established. In the case of the OSNR phenomenon, a match factor of over 0.99 was obtained. The conducted research shows large possibilities of using CNN to simultaneously monitor at least three phenomena and encourage to undertake research on the possibilities of simultaneous monitoring of more phenomena with the use of these networks.
The article presents the possibilities of using the Asynchronous Delay Tap Sampling and Convolutional Neural Network methods to simultaneously monitor the impairments of Chromatic Dispersion and Optical Signal to Noise Ratio. Using the ADTS method, which allows the presentation of distortions in the form of characteristics, a set of 10,000 images was generated simultaneously disturbed by the combination of CD and OSNR impairments. Next, using the convolutional algorithms of neural networks, the network learning process was carried out (using images obtained from the ADTS method) in order to obtain the best model for recognizing the occurring impairments and predicting their values. After a large number of tests, very good results were obtained ensuring a high adjustment of the models at the level of matching ratio R2 above 0.99 (and even above 0.999 for models for Chromatic Dispersion). Models with such a fit meet the requirements set for monitoring systems to recognize the value of occurring impairments within appropriate accuracy limits
The article presents an image analysis method, obtained from an asynchronous delay tap sampling (ADTS) technique, which is used for simultaneous monitoring of various impairments occurring in the physical layer of the optical network. The ADTS method enables the visualization of the optical signal in the form of characteristics (so called phase portraits) that change their shape under the influence of impairments such as chromatic dispersion, polarization mode dispersion and ASE noise. Using this method, a simulation model was built with OptSim 4.0. After the simulation study, data were obtained in the form of images that were further analyzed using the convolutional neural network algorithm. The main goal of the study was to train a convolutional neural network to recognize the selected impairment (distortion); then to test its accuracy and estimate the impairment for the selected set of test images. The input data consisted of processed binary images in the form of two-dimensional matrices, with the position of the pixel. This article focuses only on the analysis of images containing chromatic dispersion.
This paper presents a method of analysis of images obtained with the Asynchronous Delay Tap Sampling technique, which is used for simultaneous monitoring of a number of phenomena in the physical layer of an optical network. This method allows visualization of results in a form of an optical signal's waveform (characteristics depicting phase portraits). Depending on a specific phenomenon being observed (i.e.: chromatic dispersion, polarization mode dispersion and ASE noise), the shape of the waveform changes. Herein presented original waveforms were acquired utilizing the OptSim 4.0 simulation package. After specific simulation testing, the obtained numerical data was transformed into an image form, that was further subjected to the analysis using authors' custom algorithms. These algorithms utilize various pixel operations and creation of reports each image might be characterized with. Each individual report shows the number of black pixels being present in the specific image segment. Afterwards, generated reports are compared with each other, across the original-impaired relationship. The differential report is created which consists of a "binary key" that shows the increase in the number of pixels in each particular segment. The ultimate aim of this work is to find the correlation between the generated binary keys and the analyzed common phenomenon being observed, allowing identification of the type of interference occurring. In the further course of the work it is evitable to determine their respective values. The presented work delivers the first objective - the ability to recognize interference.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.