Broadband over power lines: can it deliver on its promise in rural America?

The issue of a "digital divide' between urban and rural America is growing more acute as more parts of our lives are tied to the Internet and the vast array of communications its carries. High-speed, broadband connections to the Internet are proliferating in urban areas but have been slow to reach into rural areas.

If it is true, as some business observers claim, that the future well-being of communities depends in part upon a broadband Internet connection, then the effort to bring affordable broadband to rural America has critical consequences.

The majority of individual Internet users in the U.S. are still using a dial-up connection to get on-line, and some level of dial-up service is widely available throughout the country. But while dial-up Internet service, which tops out at 56,000 bits per second (or 56 Kbps), is acceptable for sending and receiving emails and other small packets of data, it is becoming glacially slow for the growing number of people who using the Internet for more than electronic letters.

Broadband is preferred by individuals using the Internet to purchase goods from on-line catalogs, download music, send photos to family and friends, research school papers, take a distance learning class, access news and weather, and pay bills. It is mandatory for rural doctors transmitting medical information, and for businesses and institutions that must move a lot of information quickly.

Broadband technologies--characterized as 'always on' and at least four times as fast as dial-up, require additional equipment and expense, even to make use of the relatively large, unused bandwidth that can be found on the telephone's copper wire. Coaxial or fiber optic cable can carry broadband, as can a wireless signal beamed from 'Wi-Fi' devices, microwave towers or from satellites. Broadband services by all these means are increasing in reach and falling in price.

Unfortunately, the most popular means of moving the broadband signals are confined by physical limitations to densely settled areas. At this point in time, there is no commercial broadband service that is effectively reaching large areas of rural America.

Two technologies that should reach commercial status in 2004 hold promise. One is the Ka band satellite wireless being marketed by the National Rural Telecommunications Cooperative (NRTC). The second are a group of competing technologies that move the broadband signal over power lines. In the past 18 months, field trials of broadband over power line (BPL) technologies in the U.S. have generated an increasing buzz in the marketplace. Because the utility infrastructure already exists to carry the broadband signal to virtually every computer in the U.S., proponents including Michael Powell, chairman of the Federal Communications Commission, suggest that a solution to the rural-urban 'digital divide' may be at hand.

IS BROADBAND 'THE NEXT GREATEST THING'?

Technology experts and business strategists suggest that the stakes are high. TechNet, a consortium of CEOs from leading technology companies such as Microsoft, Cisco Systems, Palm and Intel, in 2002 began lobbying in Washington for assistance in meeting its goal of connecting 100 million homes with broadband speeds of 100 megabytes per second (Mbps) by 2010. (To put this challenge into context, 100 Mbps is more than 50 times the top speed of the broadband signals that currently reach approximately 40 million homes, according-to Jupiter Media Research.)

Why so fast? The CEOs of the Tech Net group point out that those are the speeds the U.S. public will need to transmit and receive the sophisticated and enormous data packets needed to meet entertainment and business needs. The CEOS claim that this level of investment in broadband will be worth $500 billion to the U.S. economy.

To meet its goal, TechNet has lobbied not for funds, but for an ease in regulation. There is one glaring exception: TechNet admits that 10 to 15 million homes in the inner cities and rural America will be left out of its ambitious plan, unless some form of subsidy is applied to bridge the gap in those areas between cost of service and ability to pay

To illustrate the fate technology experts predict for a 21st Century community living without access to broadband, comparisons are made to communities bypassed by the railroad in the late 19th Century or by the interstate highway system in the 20th century. In the near future, a community without broadband will be hard-pressed to compete in a digitally driven society.

Steve Collier, vice president for emerging technologies at NRTC, says that "providing access to broadband in the 21st Century can be compared to the fundamental impact that providing electricity made in rural America in the 20th century. It's just that important for the well being and the quality of life of our rural communities."

HOW DOES BPL STACK UP?

NRTC and the Cooperative Research Network (CRN) of NRECA have teamed up to conduct three studies of power line carrier--using power lines to move data--in the past six years. In the first two studies, the feasible use of the power line for communications was restricted to low-speed, narrow-band applications such as SCADA and automated meter reading. Broadband over power line was then unproven and the technical and business obstacles daunting.

In the recent 2003 study, however, NRTC and CRN find evidence of improvements in the technology, and a maturation in the companies that produce BPL devices. The study also documents that experience in new field trials, though very limited, proves for the first time that it is technically feasible to deliver broadband over power lines in the U.S.

The study emphasizes a cautious approach, noting that by the end of 2003 there were no truly commercial deployments of BPL in the U.S., and that BPL had not yet been field-tested in conditions approximating the typical rural electric cooperative service needs and conditions. Of the approximately 1,000 consumer connections made in pilot deployments of BPL in 2003, only 25 were on a cooperative utility system, in a subdivision served by Cullman Electric Cooperative in Alabama.

The strides made by BPL in the U.S. in 2003 were real--they just weren't very big. For every question answered or advantage verified about BPL, several remained unanswered by the end of the year, particularly for rural applications.

HOW DOES BPL WORK?

Medium and low-voltage power lines can carry a high frequency broadband signal from a point of injection utilizing another broadband carrier (most likely a fiber optic cable run to a utility substation) to the utility customer, regenerating the signal every 2,000 feet or so with a signal booster hung on the primary. There is one problem spot--the transformer. That very 'noisy' piece of electrical equipment can block the broadband signal. Manufacturers have developed three ways to get the signal past the transformer and into the house. The signal can be reconfigured to 'pierce' the transformer and continue onto the low-voltage line, or it can sidestep the transformer over a physical bridge to reach the low-voltage line. A third method is to bypass the transformer and low-voltage line altogether by shooting a wireless signal from a transceiver on the utility pole to a wireless transceiver in the house.

Once in the house, the broadband signal can be distributed to every power outlet with the use of a power line modem commonly available in electronics retail stores.

BPL has achieved a degree of commercial success in some European countries (although it has also run into serious regulatory problems) largely because of a basic difference between U.S. and European utility systems. In the U.S. there is an average of well fewer than ten consumers per transformer. In Europe, there are more than 200 consumers for every transformer, which greatly reduces the capital costs of installing (and maintaining) a BPL system.

WHAT ARE SOME OF THE NEED-TO-KNOW ISSUES SURROUNDING BPL?

One potentially large expense related to extending BPL service that does not get much attention in news reports and general reviews is the cost of connecting to the Internet "superhighway". A BPL network does not exist in isolation on a utility's grid--it must access the greater Internet system. Broadband signals move from the local Internet Service Provider to a larger regional ISP and so on--much like county roads connect to state highways and then to the interstate highway system.

A utility will need a robust connection to the telecommunications superhighway that is commensurate with the potential traffic flowing in and out of its BPL network. If there are not already microwave or fiber optic connection points at or near the utility system, these will have to be constructed. The expense of building this 'backhaul' to the Internet can kill even a pilot project, as Southern Maryland Electric Cooperative (SMECO) found in 2003.

When setting out to construct a pilot installation, SMECO discovered that even in its fast-growing service area 50 miles south of Washington, D.C., there were relatively few Internet 'injection' points near its lines. In order to operate the pilot, SMECO faced a large expense in investment in the installation of fiber optic cable to carry the data signals to and from its network.

SMECO CEO Joe Slater noted that the setback in the proposed pilot has not cancelled the co-op's interest in the potential of the technology, but rather served as a reminder of the complexity of adopting a new technology.

One of the big unknowns on the performance side is how BPL will stand up when the signal is regenerated over miles of line, as it will need to be in many rural co-op systems. Pilots through 2003 were generally limited to residential neighborhoods close to network access points, and required very few series of regeneration devices.

One of the leading BPL manufacturers offered to work with a co-op in Ohio on a pilot in a residential subdivision. But when the co-op asked instead to build the pilot out on a more typical rural feeder, the company balked, saying the technology was not ready to be tested in that kind of environment.

As with SMECO in Maryland, the setback did not ca use the co-op to lose interest in BPL, but rather to understand and accept the industry's current limitations.

Even if the BPL technology performs flawlessly once tested in more remote rural areas, a significant economic and marketing challenge remains. The business case for BPL will be put to a stiff test wherever competition from other broadband technologies exists in a cooperative's service area. The two most popular broadband services are cable modem, where the broadband signal is sent along the coaxial cable that is sending the cable television signal to homes, and digital line subscriber (DSL), which uses open bandwidth on the telephone's copper wire.

Cable modem and DSL are generally thought of as suburban and urban-based technologies, as both require sufficient population density to support a cable television business in the former case and a telephone exchange center for the latter. However, cable modem and DSL services have been pioneered even in rural towns in the Great Plains. If cable modem and DSL siphon off town and village customers, the financial burden of supporting the costs of a BPL system falls heavily on the most rural customers.

In addition, there are other options for rural customers. Currently, a broadband connection can be made through the Ku band satellite, though the upfront equipment costs are high and the speed some what limited. A more robust competition for rural consumers will come from the Ka band satellites such as NRTC's Wild Blue, expected to be available by year's end, offering satellite broadband at lower cost and faster speeds.

Other options include fixed wireless systems, some already owned and operated by electric cooperatives, that deliver high-speed Internet connections by line-of-sight signals. Fixed wireless is a better fit for large businesses and institutions than for residences. There are advocates of building out lower-cost fiber cables on the distribution system as a reliable means of delivering broadband everywhere the power line runs.

Economic modeling done for the latest CRN-NRTC study suggests that BPL might be feasible in rural electric co-op areas under the right conditions. These conditions may include monthly access fees as high as $50 per household, and take rates (the percentage of customers passed by BPL service that opt to purchase) of up to 40 percent.

As a point of comparison, today about 70 percent of U.S. homes are passed by either cable modem or DSL. With prices that typically range from about $30 to $50 a month, (but can top $70), about 20 percent of customers passed are taking the service.

Another potentially large hurdle for BPL in the U.S. is resolving the question of radio frequency interference. A similar issue halted the use of BPL in Japan and England.

Ham radio operators feel that sharing unlicensed frequency with BPL will be a disaster for both. The power line communications industry claims it has the technology to manage use of the frequency and protect signals. The FCC is considering the issue.

PUSHING THE ENVELOPE IN 2004

In 2004, a few of the investor-owned and municipal utilities that have been testing the technology over the past 18 months will embark upon initial commercial service. The City of Manassas in Virginia has a goal of signing up 15,000 paying customers by year's end. Cinergy is hoping to build upon its 2003 100-home pilot in its head quarters city of Cincinnati and offer BPL to up to 60,000 customers in 2004 for between $30 to $40 a month.

Among the new pilots this year is one at Central Virginia Electric Cooperative. In the most ambitious trial yet hosted by a co-op, CVEC is building out BPL into a 4,000-home section of its service area. Density in this area is eight consumers per mile. Greg Kelly, manager of member services and business development, says that at a price of only $30 for those members in the market pilot area, the co-op expects a minim um take rate of 20 percent and possibly as high as 50 percent.

Central Virginia says it is undertaking the pilot guided by the philosophy of providing a needed service to its rural members. A substantial portion of the cost of the Central Virginia pilot is being borne by the BPL vendor, International Broadband Electric Communications, which is supplying all of the BPL equipment used in the pilot. The coop is making initial installation at no charge to the customer, though the latter must purchase a power line modem to receive the signal within the home.

The Central Virginia pilot should give the electric co-op industry a clearer picture of the potential of BPL as a viable factor in the provision of broadband to rural America. But more testing under a variety of real-world conditions will be needed to determine whether broadband over power line can be an answer to spanning the "last (digital) mile."

Broadband Options for Homes and Small Businesses

                     SPEED            SPEED
    TYPE          (INCOMING)       (OUTGOING)           DISTANCE

DSL             Averages in        Averages in    Signal can reach from
                640 Kbps to        128 Kbps to    12,000 to 18,000 ft
                1.5 Mbps           2.3 Mbps       from telephone
                range              range          exchange office
Cable Modem     Up to 3 Mbps       Averages       Wherever cable
                                   300 Kbps       television exists
Ku Satellite    400 Kbps           128 Kbps       Unlimited with dish
Ka Satellite    1.5 Mbps           500 Kbps       Unlimited with dish
MMDS Fixed      750 Kbps to        750 Kbps       Approximately 35
Wireless        11 Mbps            to 11 Mbps     miles, line-of-sight
Fiber to the    Varies but                        Wherever fiber optic
Home            very high          Same           cable is deployed
Wi-Fi           Up to 25           Varies         Limited to short,
                Mbps                              line-of-sight
                                                  wireless transfer
Broadband       Tests as fast                     With signal re-
over Power      to much faster     Same           generation, wherever
Line            than DSL or                       power line electric
                cable in trials                   service exists

    TYPE                ADVANTAGES                   LIMITATIONS

DSL             Uses existing copper wire;    Available only in more
                price competitive; speeds     densely populated areas;
                do not vary as each user      generally not quite as
                has independent bandwidth     fast and slightly more
                (not shared); for most        expensive than cable
                users not as fast as cable    modem
                modem
Cable Modem     Most popular and similar      Available only where
                to DSL in range of cost       cable service exists;
                (as part of bundled           speed can slow as more
                service can be as low as      users share the available
                $29 month)                    bandwidth
Ku Satellite    The only current broadband    Costs high, especially
                option that can serve         for equipment; weather
                most rural customers          can affect signal; Slow
                                              outgoing signal speed
Ka Satellite    Same as Ku but at lower       Not yet available; should
                cost and higher speeds        cost much less than Ku
                                              but likely more than DSL
                                              or cable modem
MMDS Fixed      Very fast; good for           Limited by distance and
Wireless        business applications         geography; initial costs
                                              can be high
Fiber to the    Exceptionally fast and        Very expensive at present
Home            secure
Wi-Fi           Very popular for local        Very limited in reach;
                area networks; relatively     security concerns
                inexpensive
Broadband       Flexible design; basic        Only 1,000 U.S.
over Power      infrastructure already in     connections, most in
Line            place to connect any          small trials; many tech-
                customer                      nical and cost questions
                                              to be answered for
                                              saturated and rural
                                              feeders

Bob Gibson is a senior program manager for Marketing and Energy Services in NRECA's Cooperative Research Network.