Power to the surfers

Nationwide broadband web access, which has been promised for years, could finally become a reality using digital powerline technology delivered direct from electricity suppliers.

The slow roll-out of BT’s high-speed broadband network means internet access in much of the UK remains sluggish and unable to perform to its full potential.

But help could be on the horizon – courtesy of your local electricity company. This year a rash of utility firms in continental Europe are experimenting with digital powerline (DPL) technology, or powerline communications (PLC), using their electricity cables as a conduit for internet access. And commercial services are starting to be offered in several countries, including Germany, Spain, France and Scandinavia.

The technology could be a competitor to cable and digital subscriber lines (DSLs), or even leased lines for businesses. It is also seen as standing the best chance of commercial take-up of cable in areas where alternative and more established access technologies such as cable modem and DSL are not yet available. Nor is it just the utilities that are taking an interest: several companies have also released technology that lets DPL work in homes and offices, using the ring main.

The bandwidth on offer depends on the individual set-up. Some companies – DS2, for instance – claim they can get 45Mb/s, but according to technology research company Gartner Dataquest actual throughput may be only 12-18Mb/s. Most services offer about 2Mb/s, but most suppliers are also working to allow throughputs of 10-20Mb/s.

But the bandwidth is shared among all the users served by the headend gateway – a type of local server – that’s installed at the transformer. Although 250 users can be fitted on to a gateway, that would mean the bandwidth being severely cut down at peak times, so it’s more realistic to limit the figure to 20-60 users.

The idea of sending digital information down electricity wires isn’t new – it was pioneered in the UK several years ago. But early efforts were plagued by investment issues, worries about radiation, problems with interference, and a host of minor difficulties that put paid to the technology’s early promise.

Now DPL technology is making a comeback, as utilities realise that the technical problems with the system have been largely overcome, and the concept could yield new revenue for energy companies struggling for higher profits in stagnant markets. Prof Paul Brown, who received an award from the Communication Managers Association last year for his pioneering work on DPL, started thinking about the topic while studying for his PhD in the late 1980s. In 1990 he began research in earnest with regional power company Norweb. He says: ‘At that time the idea tied in well with the liberalisation of the telecoms and energy markets, as there seemed to be a synergy between utilities companies, as they deliver right to the customer.’

By 1992 Brown had succeeded in demonstrating an analogue phone service over power cables. In 1993 he demonstrated that digital communication was also possible, with data transmissions of up to 32kb/s using the CT2 standard for cordless telephones and ordinary power cables. Then in 1995 Norweb began trials in Manchester of digital voice communications over power lines.

It was quickly realised that the technology would have even greater application if used on the internet, and attention was switched from voice to the web. Norweb began discussions with network equipment vendor Nortel, and in 1997 the companies announced a joint venture called Nor.web to commercialise the technology. A pilot project was set up in a Manchester school, where eventually 12 workstations successfully connected to a system that delivered internet access via the power grid.The system worked like this: electrical utilities can transmit low-frequency signals at 50-60Hz and higher-frequency signals above 1MHz without affecting the other. The former carry power and the latter can carry data. A DPL base station was positioned at the local electricity substation, after the transformer, and connected to the internet using fibre or copper wire.

All homes on that transformer could be connected to the DPL system. In homes a coupling unit had to be installed beside the electricity meter, and specially adapted modems were connected to it. Using Time Division Multiplexing, the system could be set up to both send and receive data at transfer rates of up to 1Mb/s. All current DPL systems are essentially variations on this basic theme.

Early trials, however, showed up some problems. Power cables are designed to transport electricity with ultra-low frequencies of about 50Hz. But DPL called for signals in the frequency range 4-6MHz to be carried. This caused difficulties because, unlike coaxial cable, normally used for sending electronic signals, power cables do not have much in the way of screening. This meant the signals could easily leak out and cause interference, leading to scare stories that emergency services passing near DPL trials could have their transmissions interrupted. In trials in other parts of the world concerns were raised that aircraft communications could also be interrupted.

But Brown says these problems can be overcome by careful choice of frequency. Despite technical success, the Nor.web project was closed down in 1999. Brown blames a difference in the culture of the electricity and telecoms businesses. Whereas telecoms firms have been willing to fund large amounts of technical research, electricity companies are traditionally more reserved and will often only buy into new technology when it has been extensively proven. The investment ran out and the project was shelved.

Utilities across Europe are reviving the idea now because radiation from the power lines has been reduced, and technical trials have been successful. Gartner Dataquest said in a recent report: ‘PLC is a viable technology for data transport; field trials have consistently shown acceptable performance. The issue for PLC has changed from can it be done to should it be done?’ However, Marie Austenaa, senior consultant at Analysys Consulting, urges caution: ‘My view is that PLC is not yet fully proven, although there are a number of trials and some commercial launches across the world.’

In Europe, implementations are underway in Linz in Austria with ESG Elektrizität, France under the auspices of EdF, Finland with Sener, Sweden with Vattenfall/GEAB and Italy with Enel. Germany boasts at least five projects, involving major utilities such as MVV Energie. There are three projects in Spain, including one from Endesa, the biggest electrical utility in the country. Trials are also going on in the US, Brazil and Chile.

Prospects for the technology are poor in the US, however, because of differences in the electricity infrastructure. In Europe the power grid uses 220V transformers that each serve about 200 homes, whereas in the US 110V transformers serve only five to 10 homes. As the base station devices sit between the transformer and the home, that would probably make it uneconomic.

Even if the market for internet access by power line using DPL technology fails to take off, the technology has found another niche – home networking. Numerous companies from Ascom to start-ups like PolyTrax have produced hardware that allows the home ring main to be turned into a home computing network. The idea is simple. plug into any electrical socket around the house for an instant network connection. Most PLC devices for the home allow speeds of up to 6Mb/s.

However, Austenaa raises concerns about the safety of such technology in the home: ‘Currently there are health and safety issues to be discussed regarding the close proximity of people to possible emissions.’

Cost, however, will prove the real test and, with most of the technology in the trial stage, this has not become clear. But in Germany DPL costs a surprisingly low e15 (£9.50) a month plus e120-e150 (£75-£94) for installation. Gartner Dataquest reckons that for the technology to catch on in a big way with consumers it would need to cost less than $25 (£16) a month, with businesses having to be prepared to pay considerably more.

Only the next few years will see whether the technical promise of DPL is matched by commercial success.