Network Architectures, Options, Benefits and Considerations
28 November 2018 | Reading Time: 4 minutes
Which type of fiber network should I choose?
Network architectures are chosen early in the planning process but have a lasting business impact, so it’s important to assess several internal and external factors before planning begins. One of the key decisions to make is the choice between an active or a passive network type.
In an active optical network (AON), a point-to-point architecture is used and a “home run” fiber connection is established between the central office and the end user. One laser transceiver is placed in the central office while the other one is in the subscriber location. It covers longer distances and delivers high bandwidth to any single point. In a passive optical network (PON), a point-to-multipoint architecture uses unpowered fiber-optic splitters to serve multiple end-points. This consists of an optical line terminal (OLT) in the central office and optical network units (ONUs) or optical network terminals (ONTs) at the user location.
In an AON architecture, the costs of maintaining the required electronics are higher so this type of architecture is more expensive to operate. PON, on the other hand, reduces the amount of fiber and central office equipment required.
Types of PON
There are several types of commonly used PON technologies. This includes GPON (or Gigabit PON) which is IP-based, used in most deployments, and offers speeds of 2.4888 Gbps downstream and 1.244 Gbps upstream on mostly single fiber. 10G PON, which is also called XG-PON, based on the ITU-T G.987 standard and is designed to coexist with GPON on the same network. XGS-PON is 10 Gbps symmetrical which is an improvement on previous generations of XG-PON that offered only 10 Gbps downstream. This is ideal for today’s business services and mobile backhaul, while it is also relatively easy to scale up in response to demand.
NG-PON2 is 40 Gbps symmetrical – possibly 80G bps in future – with extremely high bandwidth, multiple wavelengths and software-defined networking allowing NG-PON2 to use a single fiber for different purposes. Indeed, NG-PON2 and GPON can share the optical distribution network (ODN), benefiting operators who combine business and residential services.
Main PON topologies
A typical PON covers an area of 20 kilometers in length. There are several different types of PON to choose from, with the possibility of placing the optical splitter at different locations in a PON-based point-to-multipoint network:
- A centralized splitting architecture uses single-stage splitters in a central hub in a star or daisy-chain topology to provide optimal flexibility in the management of subscriber connections and utility of connected equipment. Although it requires a “fiber rich” network from the splitter location to the premises, it also has the advantage of an easily accessed testing point. Used extensively to reach subscribers in initial FTTH deployments, multiple 1×32 splitters are typically located in the FDH (fiber distribution hub) which can be placed anywhere in the network.
- A distributed split (cascaded) architecture utilizes multiple splitters in series to achieve the overall desired split ratio, reducing the amount of fiber in the distribution area by moving part of the splitting process to the access point where the subscriber drops are connected. However, it’s worth noting that a cascaded PON network generally has a poorer OLT port utilization, than a centralized split architecture, as well as being highly dependent on the “take up rate” and the number of customers being fed from the PON.
- Daisy chaining can speed up deployments. A multi-fiber cable (connected through a cascade of fiber access terminals) can save on cable use and labor deployment. However, this approach may require more splicing than a star architecture as well as special splicing skills because in this topology, fiber cable is run through the streets and a hardened terminal is spliced onto the cable at each access point; splicing for centralized splitting is more costly than distributed splitting.
- A star architecture pulls cables back to a central location using pre-terminated cabling. Splicing takes place in the hub which makes it very efficient from a splicing perspective. However, it uses about 35% to 45% more cable than a daisy-chained architecture – and there can be more port numbers due to different cable lengths. Although the extra cable required in the star configuration carries additional labor costs for deployment, as well as physical space requirements, a star architecture can use a multiport service terminal (MST) – a component of pre-terminated connectivity lines. Dropped fibers don’t need to be spliced at the distribution point and each terminal tail – if brought back to a splice location – hence the name “star.”
- Distributed tap architecture uses fiber-optic taps in a linear topology, instead of splitters. The optical signal passes through the tap and continues down the fiber, while the tap “drops off” a portion of the signal for locally connected subscribers. This allows the typical PON reach (usually a 20 km radius from the OLT) to be extended. This is particularly useful in rural-type applications where housing density is low and distances are typically long. Multiple taps can be placed down the line until the optical link budget is exhausted or the maximum number of subscribers per OLT port (typically 32, though 64 or more are supported) has been reached.
- Fiber indexing uses connectorized cables and terminals, and enables installers to use a cookie-cutter approach to build out the network. In this topology, a reduced set of cable lengths are daisy chained together, limiting the need for custom cable assemblies or splicing. For example, the basic building block could be a 150-foot length of cable which is repeated throughout the service area – with a built-in splitter, pre-terminated 12-fiber inputs and outputs, and four or eight drops to the homes. Fiber indexing has the potential to reduce construction and civil works costs in the distribution network by up to 70 percent and thus reduce both deployment times and time to market. The reduced length of cable needed, made possible by changing the network topology and consolidating the functions of multiple network elements into the service terminal, is one of the key savings.
More fiber in the network gives cable and broadband operators additional bandwidth along with other important benefits such as lower operating costs, energy use, and carbon footprint. When choosing a network architecture, operators should consider the network size, intended usage, budget and flexibility requirements. As today’s network technologies advance, operators can choose from different strategies and approaches to bring fiber deeper into their network. Overall, we are seeing a shift toward designed-in flexibility and reliability – as it becomes increasingly essential for operators to be able to respond to fast-changing demands and service requirements.
You can learn more about FTTX considerations here.