Interleaved polling versus multi-thread polling for bandwidth allocation in long-reach PONs

Long-reach passive optical networks (LR-PONs) suffer from extremely long propagation delays that degrade the performance of centralized algorithms proposed for upstream bandwidth allocation in traditional PONs. This is because these algorithms are based on bandwidth negotiation messages frequently exchanged between the optical line terminal in the central office and optical network units near the users, which become seriously delayed when the network is extended causing the performance to degrade. In this paper, we review and analyze two centralized dynamic bandwidth allocation algorithms, online interleaved polling and offline multi-thread polling that was recently proposed in the literature for LR-PONs. We investigate and compare their performances together in detail, by studying and observing their elemental delays. Unexpectedly, simulation results show that, although multi-thread polling succeeds in decreasing reporting and queueing delays, interleaved polling keeps a lower grant delay and therefore has better overall delay performance. The latter also achieves better throughput compared to multi-thread polling.