1. Introduction
Driven by rapidly increasing application bandwidth demands, standardization in the field of Passive Optical Networks (PONs) –one of the leading solutions providing Fiber-to-the-Home – has moved towards the concept of Next Generation PONs (NG-PONs). This implies both downstream and upstream aggregate rates of several tens of Gbps. Up to now in all PON solutions (like e.g. [1]) downstream/upstream capacity is shared among Optical Network Units (ONUs – located at the user premises) in the time domain. This is due to the use of a simple optical power splitter/coupler device at the Remote Node (RN) which allows significant reduction in fiber deployment costs. This implies that in the downstream direction traffic is broadcast to all ONUs, with each ONU keeping only the data destined to them each time, while in the upstream ONUs are allowed to transmit only within time slots dictated by the Optical Line Termination (OLT) Medium Access Control (MAC) layer (located at the Central Office–CO) in order to avoid collisions with transmissions from other ONUs (TDMA operation). However, it is constantly reported that lots of practical issues emerge when trying to scale up to higher capacities, mainly due to the need of very high-speed optical components and burst-mode TDM receivers [2]. Therefore, in line with NG-PON requirements, novel technological advances which will facilitate scaling up PON capacity and flexibility are needed.