One of the great dramas in the rollout of convergence of voice, video and data services is which of the two major telecommunications networks – cable or telephone – will dominate.

The phone companies are moving to fiber-rich architectures and their approach is fairly set. The biggest question is whether to terminate the fiber at the premises or earlier in the network. The related question is how long they can maintain the copper in their network. While copper is capable of carrying voice, video and data short distances, the tricks telcos must employ to enable it to support the speed and volume demands of today’s users are growing far more complex.

The cable industry has a different set of issues. A generation ago, the industry evolved from its initial approach – evocatively called “tree and branch” – in which large coaxial trunks fed an ever-smaller network of coaxial tributaries. During the 1980s and 1990s, operators spent billions of dollars transitioning to an approach, labeled hybrid fiber/coax (HFC), in which fiber fed optical signals to local nodes, where they were transformed to radio frequency (RF) and carried on coaxial cables to customers.

Driven by increasing demand and the need to provide individualized interactive services, the industry is getting ready for the next step. The industry probably is heading to a full fiber future, but the near-term goal is to stretch the network and its protocols for the next few years. “The interim step is to bleed the … infrastructure and investment already made in equipment a little longer and then jump to true fiber-to-the-premises,” says Vince Vittore, an analyst for the Yankee Group.

Coaxial Cable: Not Dead Yet

The good news is that there is a lot of life left in coaxial cable, which the industry refers to as “coax” as if it is an old friend – which in many senses it is. Indeed, coax has tremendous capacity. The trick is to shave the number of homes served by each node to a point that interactivity, which creates bandwidth-intensive one-on-one relationship between users and networks, can be supported. “The whole idea is to make the number of people who share bandwidth on the coax smaller,” says Paul Connelly, the vice president of Business Development for Cisco’s Service Provider Video Technology Group.

The key prerequisite is vendors and engineering departments that are good at mathematics. The first key piece of the equation is the size of the serving group, which is the number of people who are served by a headend. The serving group can be supplied with signals by more than one node. For instance, a 2,000-home serving group may be comprised of four 500-home nodes. Until a few years ago, an interactive service to the four nodes – the entire 2,000 premises -- likely would be handled by a single transmitter at the headend.

No longer. The first step to combat the explosion of bandwidth demand and interactive services was to relocate the interactive gear at the nodes. The next step, which is ongoing today, is subdividing the nodes in an effort to further reduce the number of premises served. It’s tricky: The decision of using the node or the headend for transmission platforms is made on a service-by-service basis, Connelly says.

Previous Page Next Page