Evaluating power saving techniques in passive optical access networks

Abstract

Passive optical networks (PONs) are a preferred technology for implementing fiber-to-the-home networks. Though PONs minimize power consumption compared to digital subscriber loops (DSL), they still constitute a significant portion of the power consumed in the telecommunication network. Several research efforts have thus focused on minimizing power consumption in a PON network, e.g., optimal PON dimensioning, sleep modes, and designing next-generation power efficient PON candidates, like bit-interleaved PON (Bi-PON), wavelength split time and wavelength division multiplexed PON (TWDM-PON) and wavelength-switched TWDM-PON. A natural question thus arise is what are the power savings of these various mechanisms and are there some synergy gains if these efforts can be combined? In this work, we propose analytical models for evaluating the power saving potentials of optimal PON dimensioning, sleep modes, and next-generation PON candidates like Bi-PON, wavelength split and wavelength switched TWDM-PON. For optimal PON dimensioning, we consider a promised grade of service to the users, while to explore the sleep mode functionality, we consider state-of-the-art dynamic bandwidth allocation (DBA) algorithms like sleep mode aware (SMA) and hybrid sleep mode aware (HSMA). We then propose the power models to assess the power efficiency of sleep modes in combination with the optimal dimensioning. In addition, we further extend these mechanisms to the next-generation PON candidates and evaluate the power saving potentials. Furthermore, in order to show the accuracy of the proposed models, we validate all these analytical models with the simulation results.