Energy efficient DBA algorithms for TWDM-PONs

Abstract

Energy efficiency is of a vital significance in the design of next generation time and wavelength division multiplexed passive optical networks (TWDM-PONs). In this paper, we first review strategies to save energy in TWDM-PONs using the state-of-the-art dynamic bandwidth allocation (DBA) algorithms. The DBA algorithms should not only minimize energy consumption but should impose a minimal penalty on delay performance. In this context, mainly two DBA design paradigms can be exploited: offline and online. After reviewing the performance of various design paradigms, we propose an optimal algorithm, which minimizes the energy consumption at both the OLT and the ONUs, by combining the energy efficiency due to sleep modes and the load dependent use of transceivers at the OLT. Due to this, the average energy consumption is reduced to 31%.Keywords: Dynamic Bandwidth Allocation, Passive Optical Networks, and TWDM-PON. INTRODUCTIONThe incessant growth of bandwidth intensive applications like high-definition Internet video, file sharing and video conferencing requires a parallel growth in bandwidth capacities, which can be achieved by bringing an optical fibre closer to an end user, leading to the fibre-to-the-home (FTTH) technology. A promising and widely adopted FTTH technology is a passive optical network (PON). The PON mostly has a tree topology and consists of an optical line terminal (OLT) in a central office, passive splitters/combiners in a remote node (RN) and a number of optical network units (ONU) at a user side. The PON technology can be implemented in a variety of ways, and time and wavelength division multiplexed PON (TWDM-PON) has been adopted as the primary candidate of the next generation PON solution [1]. TWDM-PON combines the expansion in capacity deposited by WDM with the inherent resource granularity of a TDM PON. TWDM-PON requires a dynamic bandwidth allocation (DBA) algorithm that dynamically assigns wavelengths and time slots to ONUs. The primary objective of a DBA algorithm is to assure a high quality of service (QoS) [2]. Together with the objective of ensuring a high QoS, the DBA algorithms should also minimize the energy consumption of TWDMPONs, as current access networks (including customer's premises equipment) consumes about 80% of the energy consumed in the Internet. In this paper, we focus on the energy efficient DBA algorithm for TWDMPONs.A typical configuration of TWDM-PON has been shown in Figure 1. We assume four wavelengths at the OLT with an upstream line rate of 2.5 Gb/s and a downstream line rate of 10 Gb/s, and a split of 1:512. Further, we assume a power splitter at the RN.

Energy Conservation in Passive Optical Networks: A Tutorial and Survey

“…This will pave the way to understand the remaining sections where we explain the PON energy saving related operational mechanisms that the standardization bodies, like IEEE, and researchers from academia have considered to date. Broadly, we can define energy saving in a PON as development of less energy consuming PON hardware equipment as well as implementing techniques and strategies to operate the PON 2009 Wavelength Reusing at ONUs in WDM-PON [201] Transition overhead reduction for ONUs [200] ALR mechanism for TDM-PONs [202] Cyclic sleep mode for ONUs in TDM-PONs [200] 2010 OLT line card sleep mode [12] 2011 Selectively turning on and off transmitter and receiver of an ONU when they have no use [207] 2012 Wavelength migration in TWDM-PON [209] 2013 QoS aware adaptive sleep interval for TDM-PON [216,217] Cyclic sleep mode in WDM-PON [215] 2014 Adaptive registration of ONUs in TWDM PON for ONU migration [219] SDN based sleep mode management in GPON [218] 2015 ONU sleep mode in TWDM-PON [66,221] 2016 2017 2018 2019 2020…”

“…Rx sleep mode [207], [217] ONU turns off its receiver only A ✓ ✓ ✓ ✓ OLT-port sleep mode [22], [221] OLT turns off its downlink wavelengths (ports) [58], [245]- [247] OLT turns off its line card(s)…”

A QoS-Aware Dynamic Bandwidth Allocation Algorithm for Passive Optical Networks with Non-Zero Laser Tuning Time

“…JTWS algorithms, on the other hand, combine the wavelength assignment with time slot allocation. JTWS is more efficient and scalable but more complex than STWS [34,35]. The work of [36] decouples the wavelength assignment from the Time-Division Multiple Access (TDMA) assignment and removes scheduling complexity by using an adaptive threshold, thus leading to a lower processing requirement.…”

Low-Latency Strategies for Service Migration in Fog Computing Enabled Cellular Networks

“…The non-migration traffic includes data generated by multi-type applications, which usually have different QoS requirements but less stringent in terms of latency, packet loss ratio, etc. These different types of the nonmigration traffic can be queued independently and scheduled with different priorities according to medium access control (MAC) protocol in Ethernet PON [14].…”