25G/50G-EPON Standard Crosses the Finish Line – Enhancing Fiber Deployments as Part of Cable’s 10G Platform

25G/50G-EPON Standard Crosses the Finish Line – Enhancing Fiber Deployments as Part of Cable’s 10G Platform

Curtis Knittle
Vice President, Wired Technologies Research and Development

Jul 23, 2020

Nobody knows the extent to which broadband speeds will continue to increase over the next 5-10 years, but service providers intend to be certain that their network solutions will be able to handle whatever is coming. With the announcement of the 10G Platform the cable industry has set a new target for future broadband speeds of at least 10 Gbps, with symmetry being a key component of that new speed target.

Whereas the majority of the cable industry continues to leverage their ubiquitous hybrid fiber coaxial (HFC) networks to provide high-speed data services to a large proportion of their subscribers, fiber to the home (FTTH) solutions that utilize passive optical networks (PON) remain an important component of their solution set. Thus, it is important for FTTH technology to continually evolve in a way that meets future bandwidth demand. I am pleased to announce PON solutions have reached a new milestone in that technology evolution with the recent approval by the Institute of Electrical and Electronics Engineers (IEEE) of the 25G/50G-EPON standard.

Key Features of 25G/50G-EPON

A key requirement for 25G/50G- EPON is it must operate over already-deployed PON infrastructure with the same split ratio and nominal 20km reach as previous generations of technology. Beyond an increase is peak capacity, there are many new features and capabilities built into this standard, some of these features are focused on the efficient use of the available capacity and reducing overhead of the media access control (MAC), while others are focused on coexisting with legacy technologies and low-cost implementations. I will focus on just a few of these enhancements to the standard.

  • Peak Capacity: As the name implies, the new EPON standard developed by the IEEE 802.3ca Task Force allows for symmetric or asymmetric operation with downstream speeds of 25 Gbps or 50 Gbps, and upstream speeds of 10 Gbps, 25 Gbps, or 50 Gbps. These peak capacities are more than capable of meeting the 10 Gbps symmetric service tier goals set forth in the 10G Platform. Achieving this increase in capacity is accomplished in two ways: (1) expanding the transmission rate to 25 Gbps per wavelength and (2) leveraging wavelength division multiplexing technology to add an additional wavelength. A benefit of this architecture is that service providers can initially deploy a single wavelength providing 25 Gbps, and then add a second wavelength to bring the total to 50 Gbps upon demand.
  • Coexistence: An important consideration for service providers is the ability to support coexistence with legacy PON technologies, specifically 10 Gbps PON. Briefly, coexistence enables reuse of existing PON infrastructure by easily adding additional capacity while avoiding the complete removal of legacy PON technology. For 25G/50G-EPON, service providers who have already deployed 10G PON solutions are able to add one or more 25 Gbps wavelengths to bring total capacity to 35 Gbps or 60 Gbps in total over the same PON infrastructure. Various coexistence scenarios are shown in the diagram below.

25G/50G-EPON Standard Crosses the Finish Line – Enhancing Fiber Deployments as Part of Cable’s 10G Platform

  • Low-Cost Implementation: The economics of any residential broadband solution are always an important consideration for service providers. For optical solutions such as PON, the optical transceiver on each end of the link is one of the more costly components. Particularly for FTTH PON solutions, it is critical to keep costs as low as possible for the customer premise equipment, the optical network unit (ONU) in the case of EPON. In this regard, 25G/50G-EPON aggressively strives to keep ONU optical component costs low in two key ways: (1) fixed wavelengths, instead of tunable wavelengths, and (2) wideband optics in O-band without dispersion compensation. In fact, in greenfield deployments, e.g. new FTTH builds unburdened with legacy PON technology, a 50 Gbps ONU can use two wideband channel sources to reduce costs instead of two narrowband channel sources or operating 50 Gbps on a single wavelength, as shown in the lower portion of the figure above.

Speed, coexistence and low-cost implementations are only three of the outstanding benefits built into the next generation of EPON. Without a doubt this version of PON technology represents the lowest cost per bit compared with any other PON technology. If you are interested in looking for future activities at CableLabs related to integration of this technology into cable networks or discovering more details about 25G/50G-EPON, please feel free to contact us to receive more information.

Learn More About PON


Debunking the Myths of Shared Networks: The Point-to-Multipoint Effect

Debunking the Myths of Shared Networks: The Point-to-Multipoint Effect Curtis Knittle

Curtis Knittle
Vice President, Wired Technologies Research and Development

Oct 11, 2016

“I don’t want to have to share a pipe. The problem with ‘cable’ is shared pipes. If my neighbor is doing a bunch of stuff over the network, I get impacted too. With fiber I get speed and no shared pipes.”

--- Entrepreneur in a focus group

The notion that subscribers connected to residential fiber networks do not “share pipes” is often misunderstood. For residential fiber networks, sharing pipes is one of the main reasons fiber to the home (FTTH) is even remotely cost-effective for service providers to deploy. But what is most surprising is the following: deploying shared network solutions has led to a more rapid increase in residential broadband speeds than otherwise would have been the case with non-shared access network solutions. I like to call this the Point-to-Multipoint Effect. In the process, sharing pipes has allowed broadband speed growth to surpass the predicted 50% compounded annual growth rate commonly known as Nielsen’s Law of Internet Bandwidth. Read on to learn more…

First, a couple of definitions:

  • A (non-shared) point-to-point (P2P) network topology is one in which there is a single dedicated connection between two endpoints. In the case of access networks, one endpoint is typically located at the hub or central office, or could be located at a remote distribution point. The other endpoint is a digital subscriber line (DSL) modem, for example, or a simple Ethernet switch, located on the customer premise. In P2P networks, the peak capacity of a link is used exclusively by only the two endpoints.
  • A (shared) point-to-multipoint (P2MP) network topology is one in which there is a single downstream transmitter and multiple access termination devices that all selectively listen to the same downstream data stream. A key characteristic with P2MP networks is the peak capacity of the network is shared between all connected endpoints. Two examples of P2MP networks are HFC and passive optical networking (PON), shown in the figure below (showing downstream transmission).


Two examples of (shared) point-to-multipoint networks: HFC and PON

The PON solution represents the most prevalent residential fiber solution in the world, primarily due to lower costs compared to P2P fiber solutions. To illustrate the sharing, referring to the diagram above, if 10G-EPON is the technology choice, each optical network unit (ONU) connected to the network transmits upstream at ~10 Gbps, but they don’t transmit simultaneously. Instead, an ONU must be scheduled by the OLT for upstream transmission to avoid collisions with other ONUs. In essence, the scheduling of ONUs results in the sharing of the 10 Gbps peak capacity.  Consequently, there is a whole lotta pipe sharing going on in PON solutions.

Do shared networks necessarily perform better or worse than non-shared networks? It depends on how performance is measured, but in one key area, residential broadband speeds, shared networks have significantly outperformed non-shared networks by a substantial amount.

A recent blog discussed Nielsen’s Law of Internet Bandwidth and how the cable industry was preparing to meet future broadband speeds with 100G-EPON. When Mr. Nielsen made his initial prediction in 1998, residential broadband access was dominated by dialup and ISDN connections, which are both P2P solutions. Indeed, for approximately the first 14 years since that initial 300 bits per second dialup connection in 1982/1983, the progression of available peak service tier bit rates followed the 50% annual growth rate prediction.

**various sources compiled by CableLabs

The release of the first DOCSIS® specifications by CableLabs in 1996 essentially represented the dawn of P2MP solutions, i.e. shared, for residential Internet connectivity. According to the data in the chart above, the tremendous rate of technology advancements resulting from the shared DOCSIS/HFC network solution, and later with the development of shared PON technologies, coupled with the relative cost-effectiveness of these solutions, has far exceeded other P2P technologies for residential broadband. While the initial growth prediction in 1998 was a 50% annual growth rate, the Point-to-Multipoint Effect increased the growth rate closer to 70% for residential Internet connectivity. The Point-to-Multipoint Effect indicates that sharing pipes for residential connectivity has provided a solution that has actually allowed residential high speed data rates to increase at a faster pace! This “sharing” trend is expected to continue with the development of Full Duplex DOCSIS and 100G-EPON, making the introduction of new services possible.  Thus, just like our parents always told us, it is good to share.

In his role as Vice President Wired Technologies at CableLabs, Curtis Knittle leads the activities which focus on cable operator integration of optical technologies in access networks. Curtis is also Chair of the 100G-EPON (IEEE 802.3ca) Task Force.

Cable Information Architecture

Passive Optical Networking – for the Next Generation

Passive Optical Networking – for the Next Generation Curtis Knittle

Curtis Knittle
Vice President, Wired Technologies Research and Development

Jul 29, 2015

Service providers invest billions of dollars in their access networks. Ideally, the deployed technology meets consumer demand for many years, allowing service providers to avoid costly upgrades before fully recovering their investments. In addition to technology longevity, service providers also like to see technology evolution, a next generation, to borrow an overused technology term, to ensure future consumer demands can be met by staying within the same technology family. Nowhere is the next generation moniker more prevalent than in the development of passive optical networking (PON) standards.

Both the ITU-T and IEEE are creating next generation PON standards. The ITU has approved two documents from their G.989 series defining Next Generation PON 2 (NG-PON2), and a third document is currently undergoing final comments. The NG-PON2 architecture settled on a time and wavelength division multiplexing (TWDM) method which stacks four wavelengths in a coordinated manner onto a single fiber, with each wavelength delivering 10 Gb/s. Total NG-PON2 TWDM-PON capacity is therefore 40 Gb/s. Since the ONU only receives one wavelength, the bit rate received by a single ONU is capped at 10 Gb/s. In such a solution, the dynamic bandwidth allocation (DBA) would likely be independent for each wavelength. Multiple vendors have reported demonstrating the TWDM-PON solution. A four-wavelength TWDM-PON is illustrated in the figure below.

Knittle Blog Photo1
40 Gb/s Time and Wavelength Division Multiplexing (TWDM) PON.

Is a 100 Gb/s Solution on the Horizon?

The IEEE 802.3 Working Group has recently formed a Study Group to develop objectives for the next generation of Ethernet Passive Optical Networking (NG-EPON). Several key technology decisions await the NG-EPON Study Group: (1) number of wavelengths, (2) bit rate per wavelength, (3) transceiver tunability, and (4) channel bonding. Faced with the same consumer demands and industry competition as other access network technologies, the NG-EPON Study Group could relatively easily define a four-wavelength, 10 Gb/s-per-wavelength PON to put it on par with NG-PON2, shown above. However, alternative solutions are being investigated by NG-EPON participants that would take PON technology multiple steps beyond, and presumably allow consumer demand to be met for many years into the future. For example, using advanced modulation techniques to provide 25 Gb/s per wavelength, combined with four multiplexed wavelengths on a single fiber, could yield the first 100 Gb/s PON solution. By incorporating channel bonding, a concept popularized by CableLabs and the cable industry in the DOCSIS® 3.0 specifications, an NG-EPON ONU would be capable of receiving one or more wavelengths, potentially receiving 50 Gb/s or more. In a channel-bonded solution the DBA will closely coordinate upstream transmissions on one or more wavelengths simultaneously. Are the optical transceivers tunable? That is another of the many important technology decisions yet to be made. A channel-bonded, time-wavelength division multiplexed PON is shown in the diagram below.

Knittle Blog Photo 2
Channel-bonded, time and wavelength division multiplexed 100 Gb/s PON.

With a keen eye on vendor implementation schedules and interoperability, the increased capacity of NG-EPON ONUs and OLTs could be staged according to a timeline that aligns with consumer demand and service provider requirements, alleviating the need to develop yet another next generation standard (next-next generation?).

These technology decisions will begin to take shape within the NG-EPON Study Group (and subsequent Task Force) starting at the September 2015 802.3 Interim meeting. Anyone with the desire to contribute is welcome to attend. Perhaps the ITU-T will also investigate these same approaches for NG-PON2, ultimately resulting in another step toward a converged optical access solution (See the OnePON blog regarding a converged optical access initiative).

In his role as Vice President Wired Technologies at CableLabs, Curtis Knittle leads the activities which focus on cable operator integration of optical technologies in access networks. Curtis is also Chair of the 100G-EPON (IEEE 802.3ca) Task Force.


Technology Implications of 2Gbps Symmetric Services

Technology Implications of 2Gbps Symmetric Services Jon Schnoor

Jon Schnoor
Lead Architect: Wired Technologies

Jun 11, 2015

Service providers and municipalities alike continue their push toward offering gigabit services over fiber networks. In fact, fiberville is a web site dedicated to listing which service providers and municipalities provide fiber solutions. Recently, Comcast significantly upped the ante by announcing a 2 Gbps symmetric service that will become available in certain locations. The services announced will be 2 Gbps downstream and 2 Gbps upstream. This is a substantial announcement due to the 2 Gbps speeds and symmetrical services which facilitate faster file uploads which is of interest to individuals who work from home, small businesses and gamers.

With all that speedy yumminess, let’s examine some of the technologies required for delivering multi-gigabit symmetrical services to homes and businesses.

Setting the Stage

When a provider deploys broadband services there is typically a peak rate, above the advertised speeds, that provides the headroom necessary for them to support the speeds and service level agreements (SLAs) associated with the service.

When the total available bandwidth is shared among multiple users, like it is in PON solutions, an unscientific but common practice is for the network to support at least twice the highest advertised rate. Specifically, to support an advertised service of N Gbps, the peak rate must provide for at least 2xN Gbps. Thus, for a 2 Gbps service the peak rate must be at least 4 Gbps to safely support the SLA using common practices. This premise allows the operator to investigate and determine the technology to use in order to support the advertised speeds. Once the technology is chosen, then the engineering work required to build out the solution may begin.

Technology Options

Let’s look at the two fiber to the home solutions that will support a 2 Gbps symmetric service today: Point-to-Point Fiber and 10 Gbps Ethernet Passive Optical Network (10G-EPON).

Point to Point Fiber: Best Performance

Point-to-point topology is a “home-run” active Ethernet fiber implementation that provides dedicated fiber from the home all the way through the access network to the headend. It is analogous to building your own personal highway from home to your office so you can get to work faster. While this solution provides the ultimate future-proof network, in terms of bandwidth, flexibility and network reach, it requires a significant amount of fiber and associated optical transceivers. Running a dedicated fiber to a residential customer premise is both complex and resource intensive due to additional fiber management and ongoing maintenance. However, it delivers the best performance to meet customer needs.

Point-to-Point Topology

10G-EPON: An Efficient 2-Gig Symmetrical Solution

While there are many flavors of Passive Optical Networks (PON), (see: OnePON), 10G-EPON with its symmetric 10 Gbps links, is the only standardized, and commercially available PON technology able to provide at least 4 Gbps peak rate to support a 2 Gbps symmetrical service level agreement. Because of its point-to-multipoint topology and passive implementation, 10G-EPON is a cost effective solution in terms of operations, fiber consolidation, and headend real estate required.

PON Topology

CableLabs has championed PON initiatives through contributions to international standards, hardware and software certification, and interoperability events. CableLabs is facilitating a common approach to provide fiber solutions that will allow for quicker and higher-scale PON deployments.


Both 10G-EPON and point-to-point fiber solutions can provide 2 Gbps symmetrical services, opening up a world of possibilities for cable operators and customers alike. From the realization of all-IP delivered services, more efficient network implementations, improved cloud services, and overall future proofing the network, 2 Gbps symmetrical fiber deployments are a reality today.

Jon Schnoor is a Senior Engineer at CableLabs.


OnePON™: Addressing the Alphabet Soup of PON

OnePON™: Addressing the Alphabet Soup of PON Curtis Knittle

Curtis Knittle
Vice President, Wired Technologies Research and Development

May 12, 2014

APON, BPON, EPON, GEPON, GPON, G.epon, NGPON1, XGPON1, XGPON2, NGEPON, NGPON2, TWDM-PON, WDMPON – did I leave any out? I’m sure I did.

The alphabetical possibilities to represent different versions of passive optical networking (PON) technologies will soon be exhausted. I’m being overly dramatic, of course, but trying to make the point that all these different versions of PON, driven by membership in two somewhat competing standards development organizations, not only causes the eyes of network engineers to glaze over, but has contributed to confusion and indecision among service providers globally.

What is a PON?

In the strictest sense, a PON is an all-fiber network which consists of only passive optical components (specifically splitters and combiners), except at the endpoints of each fiber, where there is an electrically-powered termination device – either an optical line terminal (OLT) or optical network unit (ONU). The aforementioned PON technologies also assume a point-to-multipoint topology in which downstream transmission from an OLT is received by all ONUs, but upstream transmission by an ONU is received only by the OLT.


Admittedly, not all of the acronyms above represent actual PON standards, so here is an unofficial, very short description of the meaning of each acronym:

APON (ATM PON) – The first PON technology standardized by the ITU. APON used Asynchronous Transfer Mode (ATM) cells to transport data across the PON. APONs typically provided up to 622 Mbps downstream and up to 155 Mbps upstream.

BPON (Broadband PON) – ATM is known to be an inefficient transport technology, so APON was renamed to BPON to distance the technology from ATM, even though it still used ATM for transport. Supported bit rate also increased to 1.2 Mbps downstream and 622 Mbps upstream, but most (all?) were deployed at 622 Mbps downstream and 155 Mbps upstream.

EPON (Ethernet PON) – The first PON standard developed by IEEE which uses Ethernet frames to transport data across the PON.  1G EPON provides symmetric 1 Gbps links, while 10G EPON provides symmetric 10 Gbps links. There is also a variant that provides 10 Gbps downstream and 1 Gbps upstream.

GEPON (Gigabit Ethernet PON) – Many people use GEPON to refer to EPON. Others use GEPON to refer to an early, unstandardized version of 100 Mbps PON used in Japan.

GPON (Gigabit PON) – A non-ATM based PON standard from the ITU which uses GPON Encapsulation Method (GEM) to transport Ethernet frames, ATM cells, and native TDM data across the PON. GPON provides 2.5 Gbps downstream and 1.25 Gbps upstream. ATM and TDM are de-facto obsolete, and no GPON equipment supports it, resulting in a network solution which also transports only Ethernet frames – like EPON.

G.epon (EPON with a GPON flavor) – Not to be confused with GEPON, which existed first, G.epon is an ITU term used for EPON devices which can be configured using the GPON mode of ONU configuration called ONU Management Control Interface (OMCI).

NGPON1 (Next Generation GPON) – ITU name for a family of 10 Gbps PON standards (see XGPON1 and XGPON2) – this was the first “next generation” activity in the ITU.

XGPON1 / XGPON2 (10G GPON) – XGPON1 represents an ITU standardized versions of GPON capable of 10 Gbps downstream and 2.5 Gbps upstream. XGPON2 was supposed to represent symmetric 10G bps GPON, but it was never completed.

NGEPON – This is the name for the current IEEE activity to define an EPON standard beyond 10 Gbps – the next generation of EPON.

NGPON2 – This is the name for the current ITU activity to define a family of post-10 Gbps GPON solution(s) - the second “next generation” PON effort in the ITU. The technology of choice for this new standard uses Time and Wavelength Division multiplexing – the origin of the TWDM-PON acronym.

I’m sure I missed a few hyphens in all those acronyms, but I give up! It’s definitely clear as mud, right? Actually, it is very challenging to explain all the different PON standards and terminology. The bottom line is, service providers today are forced to choose between two PON solutions that do exactly the same thing – transport Ethernet frames from one side of an optical distribution network to the other side. Judging from the NGEPON (IEEE) and NGPON2 (ITU) activities, it looks like the madness will continue and the industry will have two next generation PON standards to choose from, much to the disadvantage of both service providers and vendors.

Learning from the Wireless Industry

Once upon a time in the wireless industry there were several competing standards simultaneously operating: GSM, North American TDMA, and Qualcomm’s CDMA. In addition, around the year 2000, WiMax made an entrance. These competing standards operated for anywhere between 8 and 12 years. Around 2006 the GSM/TDMA/CDMA trifecta finally converged to become what is today the LTE wireless standard.   The interesting aspect of this technology convergence is the affect on CPE dongle pricing, as shown in the graph below.


Clearly, a single technical solution drives prices downward much quicker by achieving truly global scale. This has been a huge benefit for wireless service providers. Also beneficial for wireless device vendors, they get to focus their precious and limited development dollars on a single technical solution for the entire wireless market.

OnePON™ Initiative

Clearly the global telecommunications industry does not need two different technologies to transport Ethernet frames across a PON. Today, the IEEE and the ITU are both working on yet another generation of PON standards which will continue the alphabet soup trend and continue to bifurcate the PON market. If the wireless industry may serve as an example, this will force service providers to unnecessarily pay higher prices, and device manufacturers to spend nearly twice the amount for product development if they wish to target the entire market.

The CableLabs OnePON™ Initiative is building toward creating a single PON solution that satisfies the requirements of the global service provider market. OnePON™ is not EPON, and it is not GPON. A third standardized PON solution is not a goal of the OnePON™ Initiative. On the contrary, the goal is to combine the best aspects of both IEEE and ITU solutions to create a single next generation solution.  Service providers will reap the benefits of lower capital expenses for network deployments. Vendors will be able to focus on a single technical solution for PON which addresses the global PON market, resulting in saving a tremendous amount of development costs (compared to two PON solutions), which will help translate directly to lower costs for service providers.

As Director of Optical Technologies at CableLabs, Curtis Knittle leads the activities which focus on standardizing EPON and GPON technologies for use in cable operator networks. On a daily basis he witnesses firsthand the challenges and increased costs associated with two similar PON technologies.