To meet the growing consumer demand for a more engaging television experience, many MSOs are looking to make the transition to an all-IP infrastructure. Migrating to IP enables operators to offer a more compelling user interface (UI) and user experience (UX) to subscribers, while deploying cost-effective set-top boxes (STBs) in the home.
In addition, it allows operators to leverage a wide range of subscriber devices, such as connected TVs, gaming consoles, and streaming media players like Roku and Apple TV, dramatically lowering their capital investment in customer premise equipment (CPE).
However, as cable operators increasingly look at employing adaptive bitrate (ABR) technology, they are challenged by scaling, cost, and migration path issues.
This article identifies the major hurdles that cable operators typically encounter when transitioning to an IP architecture, providing strategies for how to best address those challenges.
Benefits of migrating to an IP infrastructure
While there are numerous benefits of migrating to an IP infrastructure, one of the top reasons is an improved UI and UX for subscribers. It’s clear that today’s user experience is directly tied to the interface. Consumers enjoy the UIs offered by Netflix, Hulu, and Apple TV because they are graphically pleasing and easy to navigate. Based on the compelling and personalized user experience offered by these types of UIs, there’s a consumer expectation for a similar experience on the cable service.
However, most of the STBs deployed in homes today come with a user guide that is rather unappealing. Part of the reason for this is the large array of older STB models that are currently deployed in the field. Every time a new user interface is designed, it must be able to support all of these old models. This makes the development cycle long and not very innovative.
IP technology enables the customer premise equipmentt (i.e. the STB) to become a Web browser, allowing operators to simply pull all of the information from the network. This enables operators to make significant changes to the look and feel of the UI by only making minimal contact with the “Web server” on the network.
Not only does an IP-based model simplify changes to the UI, it also allows the operator to send personalized content.
Under a traditional MPEG-2-based STB model, an operator pushes all of the program information to all of the STBs.
Sending personalized information is not an option, as some STBs don’t support a two-way connection. Thus, in order to offer a next-generation UI and UE, operators need an IP-capable set-top box.
Once the STB is IP-capable, the question becomes whether the operator wants to only use the IP capability for navigation and the guide or also to deliver the video service. Delivering the video service via IP offers several benefits to the operator. One is the ability to address a whole breadth of consumer electronic devices in and outside the home, including connected TVs, tablets, PCs, gaming consoles, Roku boxes, etc. Today’s subscribers want the flexibility to be able to consume video content over different devices while maintaining a consistent look and feel. This enables operators to deliver on that promise.
IP video service delivery also reduces an operator’s CAPEX. Today, the bulk of an operator’s capital expenditures is customer premise equipment (CPE), namely STBs. If operators can rely on the subscriber to buy the hardware and upgrade at his or her own pace, the CPE becomes a subscriber investment rather than an operator investment. There are also additional benefits for subscribers, including the ability to use the same remote control for navigation of the content and eliminating the clutter of cables around the TV.
A third benefit of delivering video services over IP is that operators can eliminate the use of MPEG-2 STBs in favor of MPEG-4, which offers the same video quality at half the bit rate. This is advantageous for several reasons. With an MPEG-4 IPTV STB, operators have the ability to only deliver the content that is being requested as opposed to the entire service lineup like with traditional QAM delivery. This allows operators to offer a more personalized service offering. In addition, IP multicast by nature enables the operator to send only the channels that are currently being watched by the subscribers in any given service group. Thus with IP, operators can realize the benefits of switched digital video without having to deploy a switched digital video system.
A final benefit of migrating to IP is that it allows operators to consolidate network delivery on one network. After transitioning to IP, an operator doesn’t need to support one infrastructure for MPEG-2 (traditional STBs) and another one targeted toward second screens on IP. A single IP-based infrastructure can be relied upon for all services. This is more cost-effective and easier to maintain in the long term.
Challenges of migrating to IP
The main challenge of switching to an IP-based infrastructure is that it doesn’t happen overnight. To make the transition, an operator needs to support two infrastructures for some period of time: a QAM infrastructure that uses MPEG-2 for video compression and a second silo for IP that relies on a mix of MPEG-4 and adaptive bit rate technology. This can be expensive and complicated.
Another concern is known as the Super Bowl effect. When it comes to IP, especially with live channels, ABR delivery is unicast by nature, which means that every subscriber request receives a unique stream. On the night of the Super Bowl, the majority of an operator’s subscribers will turn on their TV and tune into the same channel to watch the event live. Each subscriber will need to receive a dedicated IP session. Thus, if a lot of viewers are simultaneously tuned into one channel, the cable operator must utilize more bandwidth and network resources.
Another issue is related to the cost of DOCSIS. Today, cost of throughput of DOCSIS is approximately 20 times more expensive than the price per bit on QAM.
Therefore, it’s important that operators consider they’ll be paying a significant amount more with DOCSIS services.
In addition to these main challenges, IP is a complex ecosystem and quite different from what operators are familiar with.
There’s definitely a learning curve, both in the architecture as well as the day-to-day operation that operators will need to overcome.
Strategies for migrating to IP
There are multiple approaches available to ease the migration path to IP. The first approach is a hybrid gateway. A hybrid gateway is a relatively expensive device that connects to a traditional QAM via multiple tuners. If an operator wants to deliver services to other IP-connected devices and STBs in the home, the hybrid gateway performs the decryption and transcoding from cable-card encrypted MPEG-2 to ABR delivery within the home for each device that is used concurrently.
One advantage of a hybrid gateway approach is that it’s transparent to the network, meaning the operator is merely taking existing MPEG-2, decoding it, and doing the full translation to ABR in the home. Another advantage is that a hybrid gateway approach doesn’t require a lot of upfront investment in the network. The only investment is in the devices in the home. This allows the operator to control costs based on how many subscribers are introduced to the solution. The main drawback of a hybrid gateway approach is that the cost per subscriber is very high. It is hard to show a business case for utilizing hybrid gateways to support an operator’s entire subscriber base. Another shortcoming is that it’s not future-proof. The equipment would be expensive to replace, reducing the chances of an operator being able to support new standards and emerging technologies like HEVC compression, modern ABR protocols, new in-home protocols/applications, etc.
The second approach is a hybrid STB.
A hybrid STB is connected to both the IP world as well as the traditional QAM world, allowing operators to enjoy the benefits of both. It could still use a single physical coax connection, and apply Multimedia over Coax Alliance (MoCA) as well as QAM for both IP and traditional MPEG-2 transport. This approach is advantageous as it enables a gradual migration to IP. An operator can deliver live channels available on QAM today and VOD over IP, then start migrating live channels over to IP over time. The biggest disadvantage of this method is that second-screen devices such as tablets don’t have MPEG-2 or QAM capability.
Therefore, it only resolves issues for a portion of a subscriber’s devices.
An operator would still need to deliver live content to other devices using ABR.
This scenario also requires an operator to support both QAM and IP, increasing CPE costs.
The third method is a pure ABR approach.
Whether an operator is sending content to connected devices or IP-based STBs, ABR is used for all services. Under this approach, the operator is basically sending a unicast session on the network to each and every device. The advantages are that it’s straightforward, meaning it’s a much less complex approach than having to support multiple delivery protocols, and it offers the operator a unified solution that supports all devices. The challenges relate back to the Super Bowl effect discussed above. Because a unicast session is sent for each and every device, the operator will encounter scalability and cost issues. Having to send out so many separate unicast sessions means an operator will spend more on DOCSIS and content delivery network (CDN) resources in order to support each of those sessions.
The final approach combines the advantages of ABR and multicast. The idea is to deliver ABR files via multicast to a home cable modem, which then translates the files from multicast to HTTP unicast.
Unlike the home gateway approach, there is no need to perform decryption or transcoding in the home. Rather, a simple conversion of IP multicast to HTTP unicast is required.
This enables all of the other devices in the home to see the content as ABR traffic while still being delivered via multicast on the last mile to the home, eliminating the Super Bowl effect. The advantages are that it’s a cost-effective, scalable, unified solution that supports all devices. The main disadvantage is that it requires multicast over DOCSIS.
Different pathways to IP
While the motivation to migrate to IP is clear, the most effective migration path is yet to be proven. Some approaches provide an easier initial migration while others take more careful planning and integration but promise a scalable path in the medium to long term.
Out of the various strategies that support transitioning to an IP-based infrastructure, the pure ABR and ABR/multicast approaches make the most sense in the North American market given the major role that connected devices are expected to play moving forward. Today’s cable TV subscribers don’t necessarily want a STB. They want to be able to view high-quality video content anytime, anywhere, on their second-screen devices.
The North American market is always an early adopter for technology. As more and more devices become available, customers will want to use them. In addition, most North American homes generally have multiple TVs. If an operator can avoid having multiple STBs in the home, there is a significant cost savings to be realized.
Whatever migration path an operator chooses, moving video delivery from QAM to DOCSIS will increase the operator’s CMTS capacity. This issue is being addressed by Converged Cable Access Platforms (CCAP), which offer high-density universal EdgeQAM capabilities and an easy upgrade path to future integrated CMTS capabilities. By providing cable operators with an easy, flexible upgrade path towards full CCAP and an all-IP infrastructure, while minimizing operational expenses, power requirements, and rack space requirements, CCAP systems will be a key component in the transition to IP. ■
The main challenge of switching to an IP-based infrastructure is that it doesn’t happen overnight. To make the transition, an operator needs to support two infrastructures for some period of time: a QAM infrastructure that uses MPEG-2 for video compression and a second silo for IP that relies on a mix of MPEG-4 and adaptive bit rate technology.