Thanks for the article Paul.
I think it's appropriate enough I'm copying it here.
Broadband Wireless Solutions for Global Business
With changes in spectrum availability and the development of faster microprocessors, broadband wireless technology is now becoming a viable option for operators and new entrants seeking infrastructure connection to ‘the last mile'.
Suresh Arora and Alika Nagpaul
--------------------------------------------------------------------------------
In an age of economic deregulation, corporations are becoming increasingly multinational and accelerating the overall process of globalisation. Electronic commerce is changing the competitive landscape, and demand for access to the Internet is creating havoc with the existing telecoms infrastructure, which was designed for basic telephony service. In this age of information, communications infrastructure is not only a basic necessity, it is the ticket to the competitive advantage. When currency amounts, almost equal to the total GDP of a major country, move around the globe on the information highway every night, broadband networks are needed to connect businesses end-to-end.
The competitive forces unleashed by deregulation of the long-distance market throughout the world have brought about a dramatic change in the long-haul infrastructure. With advances in fibre optics, the cost of carrying bits has plummeted. And with constant innovations in ATM and IP switching, the cost of switching traffic has been tracking a similar trend.
However, the last mile is often not capable of meeting this growing demand for additional bandwidth. Two main obstacles have remained -- lack of competition and the prohibitive cost and time involved in installing traditional infrastructure. That is changing as national markets open up to foreign competition. With the World Telecommunication Organization (WTO) agreement in place, various countries have committed to, and will begin opening up their telecom markets by 2000-2002.
Sources of Fibre
Much of the long distance infrastructure has been built around fibre. The cost of putting in this infrastructure was eased by ready availability of some form of right of way, e.g., gas pipes, railroad tracks, etc. For a new entrant in the local market, it would seem that the ideal infrastructure to deliver the broadband services that businesses need is fibre. Indeed, operators are laying fibre within the metropolitan area, finding rights of way to build fibre rings around cities. However, in the local loop this right of way typically is harder to come by and does not touch the end points, i.e., residences and commercial establishments. The challenge remains as to how to get to the end points -- the customers within the buildings -- this ‘last mile access'.
The US is a case in point. While the competitive local exchange carriers in the US have built out over 50,000 network route miles of fibre, they have fewer than 25,000 buildings on fibre. Out of 1.1 million commercial buildings in the US greater than 10,000 square feet, fibre service achieves only three per cent penetration. The reason is simple. The economics of digging up the street or curb, and the lead time involved in addressing the local ordinances, make it a daunting challenge. When added to the cost of the communications equipment on both ends, and the cost of bringing fibre near the building, there are not too many buildings with enough traffic/revenue to justify this level of financial investment. Similar situations exist in every major city. Narrow streets and crowded urban centres, that also contain historical landmarks, make cable and fibre-based solutions difficult.
Wireless Pipes
By using broadband wireless technologies, one can extend fibre-quality service directly to the customer's premises. Broadband wireless access enlarges the fibre service area and provides the most economical means of doing so. Wireless technology has the following benefits:
-lower deployment costs;
-minimal disruption to the community and the environment;
-and rapid deployment of service, which allows for a faster inflow of revenue.
In addition, costs shift from fixed to variable components. For traditional wireline systems, most of the capital investment is in the infrastructure; for wireless, more of the investment is shifted to customer-premise equipment, meaning an operator spends dollars only when a revenue-paying customer signs on. There is no stranded capital since when a customer relocates or decides to terminate service, the system is redeployable elsewhere.
There are two major trends in recent years that make the broadband wireless market a reality today. The first is that spectrum in millimetre wave bands is being made available for commercial applications. Spectrum in the 24-26 GHz is available across Europe for broadband services. In the US market, the Federal Communications Commission (FCC) has auctioned spectrum in the 28 and 31 GHz (LMDS) bands. Other regions are presently watching the US and Europe and will begin to finalise frequency allocation according to various international organisations. The regulators are setting aside bandwidth blocks of between 500 MHz to 2 GHz bandwidth, ensuring ample capacity for broadband services.
At the same time, advances in technology have made wireless technology a reality. Today there are faster microprocessors and ASICs allowing for the more robust and complex modulation needed for wireless applications. Advances in technology have also helped bring the price points for broadband wireless products to a commercial reality.
The broadband systems being introduced today are point-to-multipoint. Multiplexing radios (links) allow one to share the spectrum and, hence, the associated capacity among many radios on an as-needed basis, thereby providing improved economies throughout the network. Within each sector, a point-to-multipoint system aggregates the capacity needs of its customers. Overall, with a point-to-multipoint network, the capital burden is distributed across all utilised capacity resulting in lower capital expenditures as compared to other last-mile technologies.
Key Issues
It is appropriate to examine the variables a new operator must take into consideration when building an infrastructure. The first major decision to make is the overall network architecture. To a large extent this will be driven by the business focus. Is the plan to be voice or data centric? Eventually, to compete successfully, operators will have to offer both. Businesses are demanding more and more bandwidth primarily to satisfy the needs of their data traffic. Since the need for bandwidth for data is bursty, businesses want to pay for it only when they use it. This presents a challenge as well as an opportunity. The incumbents, with their present infrastructure, are not able to deliver services such as bandwidth on demand economically. The right architecture should give them the ability to deliver bandwidth on demand at an economical cost. A network architecture utilising TDMA techniques delivers an ideal solution for this requirement.
A new operator also has to determine a coverage region that will offer satisfactory service to as many businesses as possible. In order to capitalise on the advantage of speed of provisioning, operators have to invest in the number of hubs necessary to provide coverage for the buildings being addressed. At millimetre frequencies, the coverage area for a hub will be in the one to five km radius as a function of capacity, which means a substantial up-front investment in hubs. In metropolitan areas, this can mean up to 50 hubs may be required. The right architecture, once again, should provide an optimal economic balance between the customer premise equipment (remote terminal) and the hub (hub terminal) costs. In light of the up-front capital outlays involved, how well the overall network deployment costs scale as a function of number of buildings within the coverage area of a hub and the number of users within a building, is a very important consideration to take into account.
While the cost of equipment is an important factor for an operator to consider in the overall business case, the total life cycle costs will be dominated by other factors. Roof rights, for example, have turned into a real estate industry. The fight for space in the telecom closet has just begun, and every square foot of space will command a premium. In-building wiring poses another challenge for the new operators, as it may not be suitable for delivery of broadband services. The most dominant recurring cost, as in any other network, will be the operations and maintenance costs. Wireless solutions contain an inherent simplicity in their operations and maintenance costs. The key is not to look at the radio solution as just a set of boxes, but to view it as a network. A point-to-multipoint radio network can deliver all of the differentiated services and the economic advantages only if it is viewed and managed as the heart of the access network. Unmanned hub sites and the ability to provision from a central network operations centre (NOC) are mandatory requirements.
Flexible Solutions
In today's world where technology is changing at an unprecedented pace, there are many choices available. It is also very difficult to separate fact from hype. Before we have ATM fully deployed, there is IP switching with which to deal. For a new operator who is looking to build infrastructure for the next decade, some very difficult decisions need to be made. Should it be TDM, ATM or IP? The future is data, but what about voice? Is ATM ready for the big time? The end points for voice, “good old telephone sets”, continue to be TDM and are not about to change overnight. For a new operator, all of this boils down to a huge bet -- which technology to deploy? A flexible architecture will mitigate some of these risks for the operators.
First generation products being delivered in the market today are basically the traditional wireline products wrapped around a radio transport. For broadband wireless to deliver on its promise, i.e. provide the new operators with an economic advantage and the ability to deliver a differentiated set of services, a new bottoms up approach is needed. The approach in second generation broadband wireless networks is to design the radio network from the bottom up as an extension of the fibre network, and to absorb everything from the access point forward.
This type of product development has successfully been used in the VSAT (very small aperture terminal) market. VSATs, at a cost of US$ 100,000 only 15 years ago, provided connectivity where there were no wireline solutions available. Today, VSATs compete successfully with US$ 20 per month dial-up connections. This achievement has been due to the integration of different functionality into the radio, e.g. multiplexing/demultiplexing, a network view to operations and maintenance, and design of products and services to address the cost of rooftop installations. The need for flexibility of design through incorporating the use of TDMA and ATM architectures, and the need for modularity to fit specific business requirements, has also been addressed.
These challenges, and the decisions to be made, present a daunting task for the new operators. The good news is that broadband wireless is a reality today and offers a unique opportunity for new entrants to compete with differentiated services at favourable economics. Second generation products address the challenges facing the new operators; start-up costs, future-proof architecture that helps take the risk out of deciding on one over the other, and flexibility to help mitigate the inherent risks in any business plans in terms of what type and mix of services will be successful.
Suresh Arora is vice president of broadband networks, Hughes Network Systems. Alika Nagpaul is its marketing manager.
|
