Optimizing the Pump Power for Raman Amplifier by using Artificial Neural Network
碩士 === 國立臺北科技大學 === 光電工程系 === 107 === The main purpose of this thesis is to study optimizing the pump power for raman amplifier by using artificial neural network. The data collected from the experiment is used to build a database and is applied to the Artificial Neural Network. In the experiment, w...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2019
|
Online Access: | http://ndltd.ncl.edu.tw/handle/7385wm |
id |
ndltd-TW-107TIT00124048 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-107TIT001240482019-11-07T03:39:37Z http://ndltd.ncl.edu.tw/handle/7385wm Optimizing the Pump Power for Raman Amplifier by using Artificial Neural Network 人工神經網路於優化拉曼放大器泵浦功率之應用 LU, HUNG-KAI 呂弘凱 碩士 國立臺北科技大學 光電工程系 107 The main purpose of this thesis is to study optimizing the pump power for raman amplifier by using artificial neural network. The data collected from the experiment is used to build a database and is applied to the Artificial Neural Network. In the experiment, we compare two cases of inputting wavelength and gain and increasing the bandwidth. The former is for a single wavelength in each gain, resulting in two Laser Diode power output sizes in the Raman fiber amplifier. In the experiment, it is proved that the predicted power level through ANN training is the same as the actual transmission result, simulating a single user will get a target gain of optical fiber communication transmission, and the latter after adding a bandwidth, so that a range of wavelengths can reach a common target gain. The results of the above studies are good for machine learning applied to the development of fiber-optic communication systems. PENG, PENG-CHUN LIN, JIA-HONG 彭朋群 林家弘 2019 學位論文 ; thesis 77 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺北科技大學 === 光電工程系 === 107 === The main purpose of this thesis is to study optimizing the pump power for raman amplifier by using artificial neural network. The data collected from the experiment is used to build a database and is applied to the Artificial Neural Network. In the experiment, we compare two cases of inputting wavelength and gain and increasing the bandwidth. The former is for a single wavelength in each gain, resulting in two Laser Diode power output sizes in the Raman fiber amplifier. In the experiment, it is proved that the predicted power level through ANN training is the same as the actual transmission result, simulating a single user will get a target gain of optical fiber communication transmission, and the latter after adding a bandwidth, so that a range of wavelengths can reach a common target gain. The results of the above studies are good for machine learning applied to the development of fiber-optic communication systems.
|
author2 |
PENG, PENG-CHUN |
author_facet |
PENG, PENG-CHUN LU, HUNG-KAI 呂弘凱 |
author |
LU, HUNG-KAI 呂弘凱 |
spellingShingle |
LU, HUNG-KAI 呂弘凱 Optimizing the Pump Power for Raman Amplifier by using Artificial Neural Network |
author_sort |
LU, HUNG-KAI |
title |
Optimizing the Pump Power for Raman Amplifier by using Artificial Neural Network |
title_short |
Optimizing the Pump Power for Raman Amplifier by using Artificial Neural Network |
title_full |
Optimizing the Pump Power for Raman Amplifier by using Artificial Neural Network |
title_fullStr |
Optimizing the Pump Power for Raman Amplifier by using Artificial Neural Network |
title_full_unstemmed |
Optimizing the Pump Power for Raman Amplifier by using Artificial Neural Network |
title_sort |
optimizing the pump power for raman amplifier by using artificial neural network |
publishDate |
2019 |
url |
http://ndltd.ncl.edu.tw/handle/7385wm |
work_keys_str_mv |
AT luhungkai optimizingthepumppowerforramanamplifierbyusingartificialneuralnetwork AT lǚhóngkǎi optimizingthepumppowerforramanamplifierbyusingartificialneuralnetwork AT luhungkai réngōngshénjīngwǎnglùyúyōuhuàlāmànfàngdàqìbèngpǔgōnglǜzhīyīngyòng AT lǚhóngkǎi réngōngshénjīngwǎnglùyúyōuhuàlāmànfàngdàqìbèngpǔgōnglǜzhīyīngyòng |
_version_ |
1719287837934747648 |