Fitting Scheduling Timing&#x2013;Elastic Weighted Granting (FST&#x2013;EWG): An EPON DBA Algorithm

Chuan-Ching Sue; Wei-Nung Sung

doi:10.1364/JOCN.4.000468

Journal of Optical Communications and Networking
Vol. 4,
Issue 6,
pp. 468-479
(2012)
•https://doi.org/10.1364/JOCN.4.000468

Fitting Scheduling Timing–Elastic Weighted Granting (FST–EWG): An EPON DBA Algorithm

Chuan-Ching Sue and Wei-Nung Sung

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Chuan-Ching Sue and Wei-Nung Sung, "Fitting Scheduling Timing–Elastic Weighted Granting (FST–EWG): An EPON DBA Algorithm," J. Opt. Commun. Netw. 4, 468-479 (2012)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Abstract

Online scheduling algorithms incur no idle time problem but with less bandwidth efficiency, while offline scheduling algorithms focus on bandwidth efficiency but with idle time overhead. Although a just-in-time (JIT) scheduling algorithm for multi-channel Ethernet passive optical networks (EPONs) has been proposed to increase the bandwidth efficiency without the idle time overhead, its applicability to single-channel EPONs is problematic. A new algorithm called fitting scheduling timing–elastic weighted granting (FST–EWG) is proposed and is suitable for single-channel EPONs. First, the FST determines the fitting timing for performing the dynamic bandwidth allocation scheduling algorithm in an attempt to consider as many optical network unit (ONU) demands as possible and to avoid the unnecessary idle time as much as possible. Then the EWG can efficiently re-allocate the excess bandwidth of light-load ONUs to all heavy-load ONUs. The computation complexity of the proposed FST–EWG is linear and is denoted as $O (N)$ , where N is the number of ONUs. Simulation results show that FST–EWG efficiently allocates bandwidth compared to previously proposed algorithms in terms of average packet delay and packet dropping probability, especially for higher load. In addition, by adding the non-strict priority intra-ONU scheduling, FST–EWG also outperforms the previously proposed DBAM method in terms of EF average delay and EF delay variation.

Full Article | PDF Article

More Like This

Generic QoS-Aware Interleaved Dynamic Bandwidth Allocation in Scalable EPONs

I-Shyan Hwang, Jhong-Yue Lee, K. Robert Lai, and Andrew Tanny Liem
J. Opt. Commun. Netw. 4(2) 99-107 (2012)

SLA-Aware Dynamic Bandwidth Allocation for QoS in EPONs

Su-il Choi and Jaehyung Park
J. Opt. Commun. Netw. 2(9) 773-781 (2010)

Intelligent Dynamic Bandwidth Allocation Algorithm in Upstream EPONs

Nurul Asyikin Mohd Radzi, Norashidah Md. Din, Mohammed Hayder Al-Mansoori, Intan Shafinaz Mustafa, and Sajaa Kh. Sadon
J. Opt. Commun. Netw. 2(3) 148-158 (2010)

Previous Article Next Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (13)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (2)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (15)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Symbol	Description
$T_{REPORT}^{computation}$	The computation time required for the OLT to calculate the bandwidth granted to each ONU based on the report message
$T_{\max}^{cycle}$	The maximum cycle time between the data transmissions from an ONU
$T^{DBA}$	The total execution time required for the OLT to execute the dynamic bandwidth allocation algorithm (DBA)
$T^{GATE}$	The transmission time of a gate message
$T^{guard}$	The guard time between the data transmissions of two consecutive ONUs
$T^{REPORT}$	The transmission time of a report message
$T^{process}$	The processing time required for the ONU to receive a gate message
RTT	The round trip delay between the ONU and the OLT
$T^{idle}$	The idle time induced by $T^{DBA}$ and RTT
$T_{s}^{schedule}$	The data transmission time in the $s^{th}$ schedule determined by the proposed FST–EWG
$t_{s}^{begin}$	The beginning time of the $s^{th}$ schedule
$t_{s}^{end}$	The end time of the $s^{th}$ schedule
$t_{s}^{EST}$	The earliest time when the OLT can execute DBA for the $s^{th}$ schedule
$t_{s}^{FST}$	The fitting time when the OLT can execute DBA for the $s^{th}$ schedule
$t_{s}^{LST}$	The latest time when the OLT can execute DBA for the $s^{th}$ schedule
$t_{i}^{REPORT}$	The time when the OLT receives the $i^{th}$ report message
$b_{i}^{excess}$	The excess bandwidth for the $i^{th}$ report message after the OLT determines the allocated bandwidth
$b_{s}^{excess}$	The total excess bandwidth for the $s^{th}$ schedule
$b_{i}^{request}$	The bandwidth request size of the $i^{th}$ report message received by the OLT
$b_{i}^{grant}$	The granted bandwidth size of the $i^{th}$ report message by the OLT
$b_{i}^{guaranteed}$	The guaranteed bandwidth size of the $i^{th}$ report message
$b_{cycle}^{grant}$	The maximum granted bandwidth in the designated maximum cycle time
C	The upstream rate in the EPON (1 Gbps)
N	The number of ONUs in the EPON
$F_{N}^{lastest}$	The set of the latest N report messages received by the OLT
$F_{s}^{schedule}$	The set of report messages received by the OLT in the $s^{th}$ schedule
$\|F_{s}^{schedule}\|$	The size of the set $F_{s}^{schedule}$
$w_{i}$	The weight of the $i^{th}$ report message, which is accordingly the weight of the ONU that transmits the $i^{th}$ report message

FST–EWG improvement	IPACT–Limited	IEA–EDSA1	DPA–EDSA1
Average packet delay	8%–27%	8%–24%	11%–18%
Packet dropping probability	$2.3 \times 1 0^{- 3}$	$1.2 \times 1 0^{- 3}$	$9.3 \times 1 0^{- 4}$
QoS improvement	FST–EWG versus DBAM (under load of 0.95)
EF packet delay	128%
AF packet delay	256%
EF delay variation	166%

Symbol	Description
$T_{REPORT}^{computation}$	The computation time required for the OLT to calculate the bandwidth granted to each ONU based on the report message
$T_{\max}^{cycle}$	The maximum cycle time between the data transmissions from an ONU
$T^{DBA}$	The total execution time required for the OLT to execute the dynamic bandwidth allocation algorithm (DBA)
$T^{GATE}$	The transmission time of a gate message
$T^{guard}$	The guard time between the data transmissions of two consecutive ONUs
$T^{REPORT}$	The transmission time of a report message
$T^{process}$	The processing time required for the ONU to receive a gate message
RTT	The round trip delay between the ONU and the OLT
$T^{idle}$	The idle time induced by $T^{DBA}$ and RTT
$T_{s}^{schedule}$	The data transmission time in the $s^{th}$ schedule determined by the proposed FST–EWG
$t_{s}^{begin}$	The beginning time of the $s^{th}$ schedule
$t_{s}^{end}$	The end time of the $s^{th}$ schedule
$t_{s}^{EST}$	The earliest time when the OLT can execute DBA for the $s^{th}$ schedule
$t_{s}^{FST}$	The fitting time when the OLT can execute DBA for the $s^{th}$ schedule
$t_{s}^{LST}$	The latest time when the OLT can execute DBA for the $s^{th}$ schedule
$t_{i}^{REPORT}$	The time when the OLT receives the $i^{th}$ report message
$b_{i}^{excess}$	The excess bandwidth for the $i^{th}$ report message after the OLT determines the allocated bandwidth
$b_{s}^{excess}$	The total excess bandwidth for the $s^{th}$ schedule
$b_{i}^{request}$	The bandwidth request size of the $i^{th}$ report message received by the OLT
$b_{i}^{grant}$	The granted bandwidth size of the $i^{th}$ report message by the OLT
$b_{i}^{guaranteed}$	The guaranteed bandwidth size of the $i^{th}$ report message
$b_{cycle}^{grant}$	The maximum granted bandwidth in the designated maximum cycle time
C	The upstream rate in the EPON (1 Gbps)
N	The number of ONUs in the EPON
$F_{N}^{lastest}$	The set of the latest N report messages received by the OLT
$F_{s}^{schedule}$	The set of report messages received by the OLT in the $s^{th}$ schedule
$\|F_{s}^{schedule}\|$	The size of the set $F_{s}^{schedule}$
$w_{i}$	The weight of the $i^{th}$ report message, which is accordingly the weight of the ONU that transmits the $i^{th}$ report message

FST–EWG improvement	IPACT–Limited	IEA–EDSA1	DPA–EDSA1
Average packet delay	8%–27%	8%–24%	11%–18%
Packet dropping probability	$2.3 \times 1 0^{- 3}$	$1.2 \times 1 0^{- 3}$	$9.3 \times 1 0^{- 4}$
QoS improvement	FST–EWG versus DBAM (under load of 0.95)
EF packet delay	128%
AF packet delay	256%
EF delay variation	166%

Abstract

Cited By

Figures (13)

Tables (2)

Equations (15)

Journal of Optical Communications and Networking