Open Access: Hybrid Fiber Coax Management
Management of Hybrid-Fiber Coax (HFC) related to open access

By: Bruce Bahlmann - Contributing Author (your feedback is important to us!)

Created: December 17, 2000

This article is the fourth article in a 5 part series on open access. This series includes: Overview, Connectivity Management, Service Management, HFC Management, and Transparency.

Hybrid Fiber Coax (HFC) management is one of three key areas of open access as described in the December 2000 Communications Technology (CT) article on open access. The other key areas are Connectivity Management and Service Management that were covered in the January 2001 and February 2001 issues of CT respectively.

HFC management concerns itself with the care and feeding of perhaps the most critical component of the broadband transport – the last mile (also called the local loop). This portion of the transport (shown in Figure 1) permits the following:

• Access for Content Providers to reach broadband customers and broadband consumer devices.
• Connectivity for Content Consumers to communicate with one another.
• Access for Content Consumers (e.g. TVs, Telephones, Computers, etc.) to reach an open market of Content Providers.

Figure 1: Broadband’s Last Mile

With so many providers and consumers crowded on either end of the broadband transport several concerns have been raised regarding its management. This article will take a closer look at what it means to manage broadband’s transport with particular focus on the management of HFC in an open access system.

Broadband transports that include twisted pair (copper), coaxial cable, fiber optics, wireless transmitters/receivers, and satellite have one thing in common. That is they all have one or more single point(s) of failure somewhere in their delivery of services to their customers. Single points of failure are necessary to keep costs down while delivering the best possible quality of service to customers. The fact that all these technologies have at least one single point of failure can lower ones expectations of the service and level the playing field on which they all compete against one another for customers.

The broadband operator compensates for these one or more single point(s) of failure with employees whose job is to oversee the health of their respective broadband transport. In this case, an employee could represent someone working in the network operation center that watches various monitoring systems, oversees events, dispatches field employees to resolve the problem, and keeps all interested parties informed about the status of events. Employees also represent service personnel whose job is to repair/replace defective lines or equipment. An event could be anything from a zero-outage service item (e.g. replacement of a redundant power supply gone bad) to a reactionary correction of some outage on the broadband transport. Manual efforts can actually benefit broadband operators, as it is cheaper and perhaps easier to train employees on how events should be handled rather than teach these same things to a computer. Primarily because employees can use common sense, experience, and if all else fails they can ask for help if they don’t know how to handle certain events. For example, a computer can’t determine that something is wrong when several calls come in from customers residing in a particular area – only an individual could relate such events to a potential outage. Computers are not so fortunate to possess this experience or common sense and as a result are left with the handling of more routine and well-known events.

Unfortunately, the benefits of manual intervention in the management of broadband can be detrimental as well. Humans, though exceptionally resourceful, are prone to make mistakes. Stress as well as inexperience can result in expensive lessons learned for both the broadband operator as well as their employees. As a result, broadband operators have become increasingly cautious about making hasty changes to their complex transports. This is not helped by the fact that these transports now carry multiple services, each with its own operational criteria. Any change to these multi-service transports to resolve one problem could in effect render the other services inoperable. No more is this as important as in the case of open access.

Open access demands another level of management beyond that which most broadband operators are equipped. Never before has the importance of knowing the operational health of their transport been so important. Yet today’s broadband operator is ill equipped to manage their transport down to the End-of-Line (EOL). The EOL is traditionally associated with copper based transport media and it defines a location where the signal terminates. EOL could be a customer home or even a junction where some number of customers could connect. Although some broadband operators have been able to deploy some EOL monitoring devices that can provide them with complete transport visibility, most have not (because of the expense). Instead, these broadband operators only monitor major portions of the transport – leaving the remaining portions (which can represent anywhere from 30-50% of their customers) invisible. The invisible portions must rely on the broadband operator’s existing (and even potential) customers to alert them of potential outages.

Relying on subscribers to report outages will be further challenged by open access. This is because many services that are created as a result of open access may not all be owned or directly distributed by the broadband operator but rather by third parties. Confusion is likely to result in these instances as if the service doesn’t work whom does the customer call (broadband operator, their Internet provider, or the service provider). Once more, how can a broadband operator manually determine something is wrong with invisible portions of their transport when customer calls go somewhere other than their call center.

The answer to this may lie in technologies that provide further visibility into a broadband operator’s transport. These technologies include non-invasive network management, EOL devices, and possibly even intelligent use of customer devices (such as CPEs, cable modems, set-top-boxes, etc.).

When a telephone company designs a phone system that provides service to a number of customers it rarely supports a case that allows every residence to use its service at the same time. Although it is theoretically possible to provide such capacity, phone systems typically don’t dedicate the resources it would require to support this case. Instead, they maintain support for a percentage (25-40%) of their customers to use the service at the same time. This percentage typically covers peek usage but not much more – again so as to conserve resources and keep down costs.

This conservation of resources is a means of managing the capacity of ones transport and everyone in the business of providing a service worries about capacity. Managing capacity allows the greatest distribution of service at the least amount of cost. Not managing capacity requires huge expense with little (if any) regard to normal usage patterns. To effectively manage capacity, one needs to continually keep tabs on the usage of the transport. Because the transport is a finite medium, it must be closely monitored to ensure that traffic does not reach capacity (even during peek times). It must also be flexible enough to increase capacity where it is needed.

The advent of Quality of Service (QoS) presents some interesting challenges for capacity management – especially in the case of open access. QoS provides dedicated capacity (or bandwidth) between a service provider and the customer. For example, QoS provides dedicated capacity for the customer to place a phone call over broadband so as to ensure that the quality of the service is not impacted by other services that compete for finite capacity on the transport. While a phone call is small in relative sense (representing only a fraction of the capacity available), the use of QoS to increasingly reserve portions of a finite resource will represent a challenge for capacity management. This is because it is unlikely that the whole broadband transport will be made available for QoS. Again, like the telephone company phone systems, broadband must maintain support for a percentage of their customers to use QoS services as well as support other services that run on the broadband (e.g. best effort type services).

In open access, the broadband operator could quickly lose touch with just what types of services are running over their broadband transport. Some services may even use the same QoS capacity that is reserved for telephone or Video on Demand (VoD) so as to ensure their service does not encounter any roadblocks during delivery. If the broadband operator does not monitor their capacity they could even face legal problems – such as the blocking of 911 calls due to limited capacity. Open access could also more quickly usher in new services that demand even more capacity. If these services push maximum capacity other services could be impacted.

Managing capacity is also of concern outside the HFC where the routing of packets can change quickly from one Internet Service Provider (ISP) to another that could go against previous capacity planning. These swings in customer choice from one ISP to another may be the result of unthinkable circumstances such as a merger or a severe problem at an ISP that suddenly forces its customers to choose a new ISP. Since these swings in choices cannot be foreseen, it is highly unlikely that the current capacity is ready for this change.

The answer to managing capacity will lie in a broadband operator’s ability to oversee their transports in terms of how much bandwidth is being used and what types of services are being used. This requires bandwidth analysis tools as well as keeping tabs on the services out there under development. Together these will provide experienced staff with the information they need to stay ahead of future changes in the need for capacity. It will also make the most optimum use of resources (e.g. hardware) so as to keep the system sufficiently sized to handle the up coming need for capacity while keeping down costs.

As more bandwidth is allocated to supporting open access (or perhaps an array of digital services) broadband operators will find themselves in need of a way to effectively manage spectrum. Today, they manage spectrum with paper and pencil – constantly conducting meetings with different business units to provide the best combination of services to allocate to the available spectrum. Most broadband operators allocate every bit of spectrum they have available to new services so as to maximize their revenue opportunities. Since these services reside on specific (static) frequencies the spectrum they consume is considered dedicated to that service.

In the digital domain several of these static things change. For example, a single digital video channel only requires about 1/12^th the bandwidth of the same channel on analog. In addition, advanced technology allows up to 16 digital video channels within the space of a single analog video channel through the use of what is called statistical multiplexing (or stat-muxing). As a result, the concept of static frequencies reserved for specific services is becoming a thing of the past. In fact, managing available spectrum could be the broadband operator’s ticket to optimizing their transport in such a way as to provide them with the capability to deliver additional services as they become available.

Figure 2: Spectrum Management

Figure 2 shows how stat-muxing could potentially allow services like video, data, and voice to be combined in the most spectrum friendly way possible. In this way, the needs for QoS with voice as well as some data services could be used with the least amount of spectrum – potentially only what they need rather than the minimum defined by the QoS.

Open access will push today’s manual spectrum management to its limits as an increasing number of ISPs and service providers will have direct access to broadband customers. Since each ISP and service provider will bring something unique to the table – these value added services will provide broadband customers with an increasing number of service options (including a la carte services). As a result, the spectrum required to run all these services will require flexibility beyond that which manual static allocation of frequencies can support.

The answer to this growing need for spectrum is automation. Essentially, open access will require dynamic spectrum allocation to provide bandwidth on-demand for the growing number of services offered to customers.

The delivery component that broadband operators are least concerned about regarding open access seems to be their HFC transport. However, their ability to manage outages, capacity, as well as spectrum is far from poised to take on the challenge of open access. If not addressed, solving these HFC problems post open access will be much more difficult exercise and could leave their customer’s questioning their reliability.

Reducing the number of homes passed on a fiber node is a way that broadband operators use to free up more bandwidth to customers. Interestingly, this doesn’t actually increase the amount of bandwidth available to customers rather it merely reduces the number of customers competing for the same bandwidth. Increasing available spectrum for services will become more attractive as the amount of bandwidth required by any small number of homes is greater than what can be currently delivered via conventional CMTSs. It very likely that costs to reduce homes passed per CMTS is a losing battle – what is needed is more spectrum not fewer homes passed.

Can Birds-Eye.Net help you or your Company? Receive your Birds-Eye.Net articles and white papers hot off the presses by adding our RSS feed to your reader.

(C) Copyright Birds-Eye.Net, All rights reserved.
It is against the law to reproduce this content or any portion of it in any form without the explicit written permission of Birds-Eye Network Services, LLC. Federal copyright law (17 USC 504) makes it illegal, punishable with fines up to $100,000 per violation plus attorney's fees.