Last week I had the opportunity to attend Lightspeed Europe 2001, held in London. The conference had a common exhibition floor but two separate paths of workshops a) Lightspeed Europe, about optical networking and b) Lastmile Europe, about broadband access.

The conference was well attended although I was expecting a far larger crowd. According to an exhibitor the attendees were fewer than the past few years but better qualified. I guess it is another sign that a lot of froth has left the market…

I mainly followed the Lastmile path because I believe this is the key obstacle to achieve the broadband-society vision that fascinates large parts of the economy.

According to the Yankee Group, nearly two-thirds of residential Internet users want high-speed access, with 40% willing to pay a minimum of $40 a month for service. As Internet penetration climbs and more broadband applications come on-line, this number is expected to increase rapidly.

High-speed networks, broadband applications and a better quality of service are required, but despite a massive cumulative investment by telecoms players on improving their services supporting broadband connections, improvements are not being made as quickly as necessary.

Recent developments in optical networking, which translate into lower costs and greater network capacity and performance are crucial, however a solution to the last mile will be the key turning point to overcome the current broadband restrictions.

Demand for broadband is increasing at an exponential rate and is expected to do so for the foreseeable future. It is thought likely to be even greater in the business market than in the consumer network, if nothing else because more product development will address the market which has more clearly defined needs, hence is willing to pay more for it (video conferencing, tele-commuting are two popular applications after September 11).

The telecommunications community saw this emerging demand for bandwidth and has responded by upgrading their core networks to handle the anticipated traffic. Many of the big players continue to lay fiber in a very aggressive fashion and the advent of Dense Wave Division Multiplexing (DWDM) has increased the capacity of the fiber even further. In this way the long haul city connections are unlikely to reach capacity anytime soon. Laying fiber is very expensive (between $90 and $250 per foot) because it is necessary to dig a trench in the ground (or use other underground channels like the sewers as presented at the conference by Citynet). According to Vertical Systems Group in the US 76% of mid-sized businesses are within 1 mile of fiber. The hot topic is how to deliver the bandwidth to these businesses without incurring the expense or without disrupting the rest of the population with continuous roadworks.

In the rush to provide broadband access to million of households or businesses three dominant technologies have emerged: cable modems, xDSL and fixed wireless. Each of these technologies has particular advantages and disadvantages. Cable modems and xDSL both benefit by leveraging an installed base of wireline infrastructure. However this wired infrastructure was never intended to provide high-speed data. Upgrading existing copper to provide broadband connectivity is expensive and maximum data rates are severely limited. Wireless technologies, on the other hand, require no installed base of infrastructure but have to tackle hurdles such as foliage, weather conditions, and possible interference from other users.

My view, from the LastMile Europe Conference, is that not a single technology is perfect for every situation; the solution will be a combination between fiber and wireless adapted to the specific need of the area. Someone once told me that if we were to provide telephones to China using the same technology used in the West, there would not be enough copper in the world! Innovation is a must.

Considering the alternatives to fiber and copper, the most important solutions are: Local Multipoint Distribution Systems (LMDS), Multipoint Multi-channel Distribution Systems (MMDS), Satellite Distribution and Free Space Optics (FSO).