High speed, pervasive, mobile data is coming. In the not too distant future you will be able to access data whenever you need it, wherever you need it and from whichever device you need it. You will be able to send email while you wait for the check after lunch. You will be able to access your corporate LAN from the back of a cab. You will be able to view real-time video surveillance of your house while you're on the road. You will have access to all this, and more, from your laptop, PDA or cell phone.

Mobility is a thirty-year trend, and we are ten years into it. Cellular pioneer Craig McCaw once remarked that wireline-telephone technology “makes absolutely no sense. It is machines dominating human beings.” McCaw believed that telephony should be about communications between individuals, not between fixed entry points to the telephony network. He envisioned a world where individuals needing to communicate while away from the home or office didn’t have to search around for a telephone booth - they could just make a call from a portable, cellular telephone. That visionary belief drove McCaw to build a nationwide cellular network to provide individual, wireless telephony anywhere across the country.

The advent of the cell phone was truly liberating. Wireless telephony untethered us from wired telephone jacks and enabled us to communicate anytime, from anywhere. Such ubiquitous access to telephony fueled the cellular explosion of the 1990s. At about the same time, the Internet entered the mainstream and we became increasingly reliant on Internet-based data services. While the cell phone was liberating, our growing reliance on the Internet has once again made us dependent of fixed, wireline connections.

Much as cellular technology unleashed ubiquitous telephony in the 1990s, a new wireless technology, IEEE 802.11, also known as Wi-Fi, promises to unleash ubiquitous data access. The concept of broadband, wireless data is not new. Indeed, cellular technologies such as 3G have long been hyped with little to show for the fuss. Wi-Fi, however, is different.

Cellular technology has always been an unlikely candidate for wide-scale delivery of high-speed data. Cellular technology invariably must trade off between coverage and capacity. Today's cellular network is designed primarily to ensure broad coverage with large cell sizes. As a result, it is unlikely that the cellular network will be capable of providing high-speed data access to more than a handful of users.

In contrast, Wi-Fi technology was designed with higher speeds in mind. Relying on smaller cell sizes ranging between hundreds to a few thousand feet (versus a few miles for cellular) Wi-Fi is capable of delivering multi-megabit speeds directly to end users. In addition, Wi-Fi also benefits from one of the fastest adoption rates seen in the technology industry. One of the key forces driving the adoption of Wi-Fi is the coming ubiquity of Wi-Fi-enabled client devices. A multitude of devices including laptops, PDAs and even cell phones are now wirelessly enabled with Wi-Fi. For the first time in the history of wireless communications, Wi-Fi presents us with an abundance of wirelessly enabled devices in search of a network to connect to.

Such adoption is, in turn, driving a rapid drop in prices. In fact, Wi-Fi prices have been dropping steadily and rapidly at a rate much faster than cell phones ever did. Just a few years ago, when Wi-Fi was first introduced, a single Wi-Fi client interface cost roughly $500. Today an equivalent interface retails for less than $50. Indeed, Intel's recently launched Centrino processor makes the marginal cost for a Wi-Fi interface on a mobile laptop is effectively zero.

Service providers such as T-Mobile and Cometa are beginning to take advantage of Wi-Fi's capabilities to offer public high-speed data access. Currently, service is available at several Wi-Fi “hot-spots” in airports, hotels and coffee shops sprinkled across major metropolitan areas. These early hot-spot deployments give us a glimpse of what a truly broadband, wireless data network could offer.

However, the hotspot model represents an incomplete solution. After all, today's incarnation of hotspots provides access that is more like wireline pay-phone access than truly ubiquitous access. Critics contend that Wi-Fi cannot scale beyond the wireless LAN extension for which it was designed.

That argument neglects to factor in the effects of rapid, market-driven innovation. As usage for wireless data increases, a solution providing perceived ubiquity becomes necessary - such was the case with cell phones and voice access.

Ultimately, Wi-Fi coverage will not be provided by payphone-like hotspots, but rather by cellular-like, city-wide hotzones. Then the question becomes how to make hot-zones work. A few challenges need to be overcome:
• Although small cell size gives Wi-Fi the opportunity to transmit and receive at high speeds, it also means that a large number of cells are required to cover a metropolitan area.
• Like every other wireless technology, Wi-Fi coverage is impeded by obstacles like buildings, trees, and terrain features. The resulting coverage area will be marked by shadows and coverage gaps unless sufficient Wi-Fi-cells are distributed to create even coverage around the obstacles.

There are a variety of possible techniques that seek to increase the range of individual Wi-Fi access points in order to reduce the overall number of cells. The principal drawback with increasing range is that fundamental limitations in propagation around obstacles remain. Obstacles will still create coverage gaps, a problem that cannot be overcome in larger cells. Expansion of the cell size is analogous to trying to provide light to a city at night using a single powerful light bulb instead of relying upon a larger number of small, well distributed street-lights. Whereas street-lights mounted on poles are capable of providing a high degree of uniform brightness throughout the city, a single more powerful lamp will undoubtedly result in unacceptable shadows and dark spots.

The key challenge to building a broad network of distributed Wi-Fi cells is that each cell requires some form of backhaul. Wired backhaul is infeasible, since building a city-scale wired LAN infrastructure would be cost-prohibitive to provision and maintain. The recurring access charges for wired backhaul alone render the economic model unworkable.

To make a hotzone cost-effective and practical requires a solution that provides the flexibility to deploy cells in a manner that overcomes any potential coverage issues without the constraints imposed by wired infrastructure.

The need of the hour is a unique, wireless network architecture that overcomes the challenges of metro-scale Wi-Fi deployments. The architecture should dramatically reduce the reliance on wired backhaul and thus enable a true broadband wireless network.

Is ubiquitous, untethered Wi-Fi coverage in our future? Only time can tell, but history suggests that people value their freedom.

As CTO, Sri led the development of the core technology underlying FHP Wireless products. Sri built the technical team at FHP. Prior to founding FHP Wireless, Sri led the development of the common network management software interface for optical transport and switching products at Sycamore Networks. He has taught courses in software engineering and mathematics at Caltech and MIT, and has received numerous awards for his work in mathematics.
Sri has an M.S. (Electrical Engineering and Computer Science) from MIT and a B.S. (Mathematics) from the California Institute of Technology.