Home broadband, video usage patterns will force changes in access, aggregation networks

Michael Kennedy


Residential networking services and the devices we use to access them are constantly changing, usually in unanticipated ways. Video streaming, social networking, smart phones and tablets are creating new demand, displacing traditional media and threatening the foundations of the multichannel video subscription (cable, DBS and telco) industry. The new services and devices also are causing a shift in video usage from multichannel service to broadband Internet service. The shift already has made video the dominant Internet traffic source and is driving network operators to rethink their network architectures.

Despite the inroads of new devices and services we still watch a lot of traditional TV. The Nielsen Company estimates that the average consumer watches about 5.5 hours per day. Ninety-nine percent of U.S. households have at least one TV and 65 percent of U.S. households have three or more TVs. Smartphones, the next most heavily used device, are used 1.2 hours per day. TV, also, is holding its own in the 16-44 age demographic--the most enthusiastic adopters of the new devices. A March 2014 study by Millward Brown found that this demographic watched TV 41 percent of the time while simultaneously using at least one other device. We watch another hour of video per day using time shifted TV, DVD/Blu Ray, Internet and smartphones.

Though heavily marketed by mobile operators, smartphones and tablets are major contributors to home broadband usage, because they are used in Wi-Fi mode in the home to reduce mobile data usage fees--Wi-Fi accounts for slightly more than half of smartphone usage time.

The rate of movement from the traditional TV broadcast (multicast) model to on-demand video streaming (unicast) whether over the Internet or via subscription is the primary driver of network bandwidth capacity.

The movement from multicast to unicast video distribution creates a massive increase in aggregation network bandwidth. For example, Sandvine reports that today 67 percent of peak period Internet traffic is due to real-time entertainment (primarily video streaming). Since we are currently watching 5.5 hours of traditional TV for every hour of Internet TV (see above), then a full shift of all TV viewing to the Internet would imply at least a 5.5 times increase in Internet video traffic. Furthermore, there is even much greater potential traffic growth because consumer electronics vendors are making it easier to use the Internet with large screen TVs. A large screen TV requires about 60 times more bandwidth than a smartphone, for example. (Small screens use fewer pixels and thus less bandwidth than large screens.)

Large screen HDTVs are continuing to gain acceptance and therefore driving bandwidth increases. Though multichannel video subscriptions peaked in the U.S. last year, digital video subscriptions, needed for HDTV, continued to climb. Also, Netflix, the leading OTT service provider, now offers 7 Mbps Super HD and 12 Mbps 3D service options that are higher speed than its 5 Mbps HD service.

Residential demand trends will force network providers to rethink the architecture of their access and aggregation networks.

Redesign of the access network should be relatively modest. Today, the majority of households subscribe to a broadband and multichannel video service bundle offered by a cable or telecom provider. The services enter the home on a twisted pair, fiber or coaxial cable. Roughly one-fourth of the total bandwidth entering the home is allocated to broadband Internet and the remainder is allocated to multichannel video. Residential usage trends suggest that the bandwidth allocation will need to shift to more heavily favor the Internet. The total per household bandwidth allocation will depend primarily on the number of TVs and their screen sizes. Personalized video on smaller screens seems to be the dominant trend and will put downward pressure on household video bandwidth requirements. However, multiscreen use and new applications such as the Internet of Things should produce new bandwidth demand that is likely to offset this downward pressure on household bandwidth requirements.

The aggregation network architecture requires fundamental rethinking. The dominance of Netflix (Nasdaq: NFLX) and YouTube traffic on today's Internet is but a forewarning of what the future holds. Operators are already moving cache, video servers and video transcoding centers closer to end users as a means of controlling aggregation transport costs. Packet Optical Transport Systems, 100 Gbps+ transport, SDN and NFV also will be used aggressively to control costs. I believe the technology is available to cost effectively build the future residential network.

There are equally big business and financial challenges looming as the dominant multichannel video subscription service model gives way to a more open Internet-based service delivery model. One of the biggest issues yet to be faced is that most advertising revenue comes from multichannel subscription services. The industry has not yet found an effective way to replace these revenues with new Internet video advertising vehicles.

Michael Kennedy is a regular FierceTelecom columnist and is Principal Analyst at ACG Research, www.acgresearch.net. He can be reached at [email protected]