Qos-aware bandwidth allocation algorithm for long range passive broadband access network

Abstract

Next generation broadband access networks are gaining more interest from many key players in this field. The demands for more extended reach and higher bandwidth are among the driving factors for such a network. The advantages of this network include a wider coverage area reaching 100 km, even beyond; increased bandwidth capacity and transmission speed, but with low cost and energy consumption. One promising candidate is the Long Reach Passive Optical Network (LR-PON). LR-PON is a term that refers to the consolidation of the metro and the access networks in the traditional PON, which also means the merger of the multiple Optical Line Terminals (OLTs) and the Central Offices (CO) in the network. LR-PON can simplify the network by reducing the number of network elements, equipment interfaces, and even nodes which significantly reduces the network’s Capital Expenditure (CAPEX) and Operational Expenditure (OPEX). Although the LR-PON can provide extended reach, this results in increased propagation delay of Dynamic Bandwidth Allocation (DBA) messages exchanged between the OLT and Optical Network Units (ONUs), leading to degradation of DBA and Quality of Service (QoS) support. Longer round-trip-time (RTT) in LR-PON causes redundancy in the DBA mechanism as the ONU polling cycle becomes smaller than RTT. Therefore, this thesis focuses on deploying an efficient QoS-aware bandwidth allocation algorithm with an appropriate Service Interval (SI) setup for LR-PON to ensure the delay is maintained under ITU-T G.987.1 standard requirement, named as Service Interval-based Bandwidth Algorithm (SIBA). OMNeT++ software is used to run a discrete event simulation of the network. Simulation findings reveal SIBA’s superior performance for all the traffic classes where the mean upstream delays of Transmission Container (T-CONT) 2 and T-CONT 3 improved up to 88% and 85% respectively, compared to Immediate Allocation Colourless Grant (IACG) algorithm. SIBA also surpasses the mean upstream delay of the Efficient Bandwidth Utilization (EBU) algorithm by up to 90% for both T-CONT 2 and T-CONT 3, and by 83% for T-CONT 4 compared to the Comprehensive Bandwidth Allocation algorithm for LR XG-PON (CBA-LR).