Multi-Thread Fitting Scheduling Timing – Elastic Weighted Granting Algorithm to Overcome Performance Degradation in Long Reach TDM-PONs
碩士 === 國立成功大學 === 資訊工程學系 === 102 === In recent years, passive optical networks (PONs) have been widely chosen for deploying optical access networks. With the widespread deployment of PONs, research focus has shifted to their scalability, with longer reach and higher split ratio. By exploiting both o...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2014
|
Online Access: | http://ndltd.ncl.edu.tw/handle/13875628714759170455 |
id |
ndltd-TW-102NCKU5392057 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-102NCKU53920572016-03-07T04:11:05Z http://ndltd.ncl.edu.tw/handle/13875628714759170455 Multi-Thread Fitting Scheduling Timing – Elastic Weighted Granting Algorithm to Overcome Performance Degradation in Long Reach TDM-PONs 於長距離分時多工被動式光接取網路中解決效能退化之多執行緒的適當排程時機-彈性權重分配演算法 Yu-TengWu 吳育騰 碩士 國立成功大學 資訊工程學系 102 In recent years, passive optical networks (PONs) have been widely chosen for deploying optical access networks. With the widespread deployment of PONs, research focus has shifted to their scalability, with longer reach and higher split ratio. By exploiting both optical amplifiers, long-reach PONs (LR-PONs) extend the coverage of a PON from short-reach 20 km to 100 km and beyond. By reducing the number of network elements, the network’s Capital Expenditure (Capex) and Operational Expenditure (Opex) can be reduced. On the other hand, due to the increased distance between optical network terminal (OLT) and optical network unit (ONU), the propagation delay increases to 1ms. When an ONU requests upstream bandwidth, it will take a longer time to receive the acknowledgement from the OLT before it can transmit its upstream packets. Thus, the traditional dynamic bandwidth allocation (DBA) in LR-PONs must be redesigned and under review. In this thesis, we propose a multi-thread fitting scheduling timing-elastic weighted granting (MT-FST-EWG) dynamic bandwidth allocation algorithm (MT-FST-EWG) which is to merge the advantages of both fitting scheduling timing-elastic weighted granting (FST-EWG) and multi-thread polling (MTP) and improve the effectiveness of FST-EWG shortage in LR-PON. We evaluate the throughput-delay performance of MT-FST-EWG, FST-EWG and MTP. We find that MT-FST-EWG gives lower delays and idle times than FST-EWG and MTP. At the same time, the propagation delay problem FST-EWG faced was solved by MT-FST-EWG. Chuan-Ching Sue 蘇銓清 2014 學位論文 ; thesis 32 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立成功大學 === 資訊工程學系 === 102 === In recent years, passive optical networks (PONs) have been widely chosen for deploying optical access networks. With the widespread deployment of PONs, research focus has shifted to their scalability, with longer reach and higher split ratio. By exploiting both optical amplifiers, long-reach PONs (LR-PONs) extend the coverage of a PON from short-reach 20 km to 100 km and beyond. By reducing the number of network elements, the network’s Capital Expenditure (Capex) and Operational Expenditure (Opex) can be reduced.
On the other hand, due to the increased distance between optical network terminal (OLT) and optical network unit (ONU), the propagation delay increases to 1ms. When an ONU requests upstream bandwidth, it will take a longer time to receive the acknowledgement from the OLT before it can transmit its upstream packets. Thus, the traditional dynamic bandwidth allocation (DBA) in LR-PONs must be redesigned and under review.
In this thesis, we propose a multi-thread fitting scheduling timing-elastic weighted granting (MT-FST-EWG) dynamic bandwidth allocation algorithm (MT-FST-EWG) which is to merge the advantages of both fitting scheduling timing-elastic weighted granting (FST-EWG) and multi-thread polling (MTP) and improve the effectiveness of FST-EWG shortage in LR-PON. We evaluate the throughput-delay performance of MT-FST-EWG, FST-EWG and MTP. We find that MT-FST-EWG gives lower delays and idle times than FST-EWG and MTP. At the same time, the propagation delay problem FST-EWG faced was solved by MT-FST-EWG.
|
author2 |
Chuan-Ching Sue |
author_facet |
Chuan-Ching Sue Yu-TengWu 吳育騰 |
author |
Yu-TengWu 吳育騰 |
spellingShingle |
Yu-TengWu 吳育騰 Multi-Thread Fitting Scheduling Timing – Elastic Weighted Granting Algorithm to Overcome Performance Degradation in Long Reach TDM-PONs |
author_sort |
Yu-TengWu |
title |
Multi-Thread Fitting Scheduling Timing – Elastic Weighted Granting Algorithm to Overcome Performance Degradation in Long Reach TDM-PONs |
title_short |
Multi-Thread Fitting Scheduling Timing – Elastic Weighted Granting Algorithm to Overcome Performance Degradation in Long Reach TDM-PONs |
title_full |
Multi-Thread Fitting Scheduling Timing – Elastic Weighted Granting Algorithm to Overcome Performance Degradation in Long Reach TDM-PONs |
title_fullStr |
Multi-Thread Fitting Scheduling Timing – Elastic Weighted Granting Algorithm to Overcome Performance Degradation in Long Reach TDM-PONs |
title_full_unstemmed |
Multi-Thread Fitting Scheduling Timing – Elastic Weighted Granting Algorithm to Overcome Performance Degradation in Long Reach TDM-PONs |
title_sort |
multi-thread fitting scheduling timing – elastic weighted granting algorithm to overcome performance degradation in long reach tdm-pons |
publishDate |
2014 |
url |
http://ndltd.ncl.edu.tw/handle/13875628714759170455 |
work_keys_str_mv |
AT yutengwu multithreadfittingschedulingtimingelasticweightedgrantingalgorithmtoovercomeperformancedegradationinlongreachtdmpons AT wúyùténg multithreadfittingschedulingtimingelasticweightedgrantingalgorithmtoovercomeperformancedegradationinlongreachtdmpons AT yutengwu yúzhǎngjùlífēnshíduōgōngbèidòngshìguāngjiēqǔwǎnglùzhōngjiějuéxiàonéngtuìhuàzhīduōzhíxíngxùdeshìdāngpáichéngshíjīdànxìngquánzhòngfēnpèiyǎnsuànfǎ AT wúyùténg yúzhǎngjùlífēnshíduōgōngbèidòngshìguāngjiēqǔwǎnglùzhōngjiějuéxiàonéngtuìhuàzhīduōzhíxíngxùdeshìdāngpáichéngshíjīdànxìngquánzhòngfēnpèiyǎnsuànfǎ |
_version_ |
1718199540942635008 |