Electronic Design

Communications Nirvana: Why Must We Strive To Reach That Plateau?

As we sew the fabric of blanket communications, reflecting on its origins may help determine where the technology is going.

These are exciting times to be an engineer. I never thought I'd see the day when industrial, financial, and pharmaceutical interests would take a back seat to electronics on the evening news. This year, from microchips to software, the NASDAQ's technology-heavy investments have nearly steamrolled all other funds in terms of returns.

Our corporate presidents and fellow engineers have taken the leap from geekdom to stardom as reporters scramble for interviews or quotes that will shed light on what companies will do next. You can't go into Starbucks for a quiet cup of coffee without hearing someone waxing lyrical about the latest virtual startup.

What's going on here? Yes, I was surprised when my friends and family started quizzing me about Windows all those years ago. It was the first time anybody outside of my field had asked me anything even remotely related to what I do for a living. Before our friend Bill Gates came along, no one knew what to ask. I learned to keep my answers short, though, after receiving only a few of those glazed looks.

Then Intel took over the show, with its cool guys in jumpsuits dancing to the latest Pentium jingle. That's my reference point when trying to figure out when everyone and their uncle began asking me about bytes, megabytes, megahertz, V.90, baud, active-matrix LCDs, digital TV, and the many other tools of the technological wordsmith.

So what happened? Why is everyone so techno-aware? Why are my morning paper and evening news becoming more and more like the tech journals and company-profile videos I thought I had left at the office—with some fluffy human-interest stories thrown in? The answer, in short, is communications. Extrapolate upon that and you get its lovechild, the Internet or World Wide Web.

But why? Why are so many companies throwing billions of dollars into shoring up the communications infrastructure? Why are we so quick to invest our hard-earned money in a self-perpetuating fashion, into the companies supplying the terabit backbones and multi-gigabit routers that make up the behemoth? That question can only be answered by looking to the near future to define what will feed the monster's huge backbone, which is currently under construction. Also, we must pin down which technologies will be used to feed it. But first, we must examine ourselves and our past.

Why Communications?
From grunts to crude symbols, hieroglyphs, elaborate texts and letters, and ultimately to photography and the moving picture, our ability to engage each other has advanced steadily through time. Then, in the 20th century, all frames of reference were suddenly lost. Steady advancement gave way to rapid technological development on an exponential level.

Now, we stand at the dawn of a new millennium with gaping excitement tempered by wary trepidation. Where is our communication technology taking us? How will it affect our day-to-day lives? Will we have a life worth affecting? All are fair questions that, to the technophobe, must be answered before we take another step. But for every technophobe or Luddite, I'll show you 50 tech-savvy entrepreneurs who are throwing caution to the wind and jumping headfirst into the fray.

The technophobe does have a point, though. The rate of change might be too rapid. While technology has advanced exponentially, we haven't. Physiologically and psychologically, we're still back in the Renaissance. Until recently, that wasn't a problem. But the rapidity of change, the information overload, the overstimulation of the modern environment, and the movement from active to sedentary lifestyles have taken their toll. Everything from odd behavior and total nervous breakdowns to medical ailments, such as premature diabetes in our obese children, has been attributed to the stresses and lifestyle changes incurred by our love of technology.

I'm convinced that it's not the technology itself, but the manner in which we interface with it that's mostly to blame for at least the stress aspects. We have a basic need to identify and categorize everything in our lives, both animate and inanimate. But the information that we base our everyday decisions on is changing so fast that the frames of reference we use to make those categorizations become blurry. Hence, we lose our ability to make the necessary decisions and "information overload" becomes the disorder of the day. This leads to everything from vague symptoms of stress to nervous breakdown, especially when the decisions that have to be made are critical to your job or the welfare of you and your family.

I can safely assert, then, that the man-machine interface and information handling will be the next killer research field. In the meantime, the question as to what will be feeding the enormous infrastructure now being built remains to be answered. From the previous discussion, you may have guessed the answer: information. Information was the hot commodity at the close of the last century. The gathering, brokering, and transportation of that information stand to become boom industries at the beginning of this century.

From a purely altruistic standpoint, our goal is to make the gathering and transportation of that information as simple and efficient as possible. Of course, in reality, our real job is to make money to feed our family. But let's not go there. Words like "altruistic" sound much more noble.

It Started With Pictures
A discussion of information must begin with the Internet, the phenomenon that has changed everything. Oddly, though, the Internet wouldn't have been possible without the efforts of our cave-dwelling ancestors. By learning to draw and communicate effectively via pictures, they set in motion a chain of evolutionary events that would shape and mold our brains. Look at the way the phrase, "A picture is worth a thousand words," takes on greater meaning as time goes by.

The Internet would never have made the strides it has if we were still typing plain text. In fact, wasn't it adult-entertainment interests that became one of the first monetary supporters of the web? Thanks to those dollars, the Internet took hold and flourished. Otherwise, it might have languished in military and academic circles.

E-mail and voice will remain at the core of our everyday communication, but they won't drive future technological innovation. It will be the conversion and transportation of high-resolution color images and video. We'll need digital video in mobile receivers, travelling in a car at 70 mph, with zero multipath distortion in built-up areas. We'll also need bandwidth for video conferencing and web casting at the drop of a hat, including high-resolution medical images.

Business is only part of it. Carriers are aiming their digital subscriber line (DSL), cable, fiber-optic, and wireless pathways right at our homes. The public switched telephone network (PSTN) won't know what hit it as it gets shoved to the sidelines by the data-intensive media knocking on our doors.

All of this can't just mean the ability to quickly download cool graphics, the latest MP3 audio file, or talk for free over iffy Internet Protocol (IP) lines. What do the carriers see in our future? What are they banking billions on as they trip over each other to get into our homes?

Yes, e-commerce is a large chunk of it. Video is there, as is audio, news, and all the paraphernalia of the modern environment. From my point of view, what's going to make it really big is universal connectivity—joining the dots. These will be the mantras of the next decade—from our thoughts to their medium, and everything in between. Wireless is where it's at, even though it will be some time before a wireless connection can compete with wireline on a performance/cost basis. Still, the convenience of an untethered connection will push many to take the cost hit as it unfolds.

It must be that built-in urge to both control our environment and to walk freely within it that has us in its grasp. We struggle to elongate our fingers by way of invisible, wireless tentacles that wrap around TVs, VCRs, wireless phones, cookers, and every other electronic and mechanical device with which we surround ourselves.

That urge, combined with a need to communicate with the outside world easily and rapidly from the comfort of our home or office, has led us to where we stand today. It will take us into the near future with a myriad of tools that are limited only by our imagination. There will be everything from connected refrigerators to handheld personal digital assistants with neural processors that "learn" about us and our habits as we use them. So what are the technologies, both wireline and wireless, that will hone these tools to the level we need to take us there?

Any examination of the varying routes to communications nirvana has to begin with the cutting edge of development: the wireless, short-range, personal-communications device. This spring will source the vast majority of information flowing into the ethereal communications cloud. Connected to everything from blood-pressure monitors to voice communicators and your toaster and refrigerator, these devices will be our personal, always-on connection to the Net.

Leading the charge into the next millennium are those devices conforming to the Bluetooth standard. Long overdue, these devices will hit the ground running as early as next month. But this will only be the beginning.

Impediments to the proliferation of Bluetooth essentially revolve around cost. With improved transmitter and receiver architectures, shrinking processor cores, and the combination of direct-conversion techniques with more sophisticated packaging, the overall cost of these devices will drop from about $20 initially to under $5 in 12 to 18 months. Then hold onto your hats, because it's going to rapidly turn into a maelstrom.

With applications outside the home ranging from corporate LANs to airports and hotel lobbies, network access using Bluetooth devices will be everywhere, tracking everything you do. Some thorny privacy issues may crop up, such as employers monitoring every move that we make within a building—ostensibly to improve communication and security. A person will be quickly located and incoming calls automatically routed to the nearest available phone. But any log kept of a person's movements could eventually be used against them.

Ironically, Bluetooth's modulation scheme—frequency-hopping spread spectrum—was chosen for its robustness and security aspects. Yes, getting information from A to B is important, but I'm more fearful of what's done with the data once it reaches its destination. Information is the currency of the new century. A proliferation of information-gathering and transmitting devices will be monitoring you, and you'll be a willing participant, eager to forsake privacy for convenience.

The convergence of the above-mentioned technological parameters with advanced CMOS processing will quickly allow low-cost Bluetooth connectivity. Operation will hit rates well above the first iteration's 1 Mbit/s.

Until wireless RF solutions reach critical mass, however, IrDA-compatible infrared devices will flourish in line-of-sight or in-room applications. These devices already enjoy the benefits of a bill of materials that's under $5. Though incapable of through-wall communication, IrDA-compliant devices will remain popular. As Bluetooth ramps up, they will leverage off their cost, relative security, and simplicity as a medium.

Application software is plentiful for IrDA-enabled devices, and many cellular phone, computer peripheral, and laptop companies have a built-in IrDA solution. My impression, though, is that manufacturers are just biding their time until Bluetooth takes hold. At that point, expect a mass migration, especially as the range of operation increases from 10 m (Class 1 and 2) to 100 m (Class 3)—the required distance for effective operation within a corporate office building.

On the home front, Bluetooth was originally thought to be competing with wireless standards such as the 10-Mbit/s HomeRF with its direct-sequence, spread-spectrum modulation scheme, and SWAP, the shared wireless access protocol (Fig. 1). But Bluetooth's 1-Mbit/s data rate makes this unlikely. Instead, the two will complement each other. HomeRF devices have a solid backing and a critical mass of enabled product lines about to hit the market. So if they don't maximize upon the head start they have over Bluetooth, they might lose their window of opportunity in the coming year. Bluetooth isn't standing still, with work already in progress for a 10-Mbit/s version by 2002.

Together, these two wireless interfaces have the upper hand over wire-based standards, which depend on a home being either wired with phone lines in every room (HomePNA) or other complete rewiring. In the past, those standards remained feasible due to the extremely wide price disparity that existed, say, between wired Ethernet setups and wireless implementations. But that contrast has weakened enough to make the argument against wireless utterly insignificant.

The next stage of connectivity deals with how the information will be shoveled from the home or office to the outside world, onto the backbone, and vice versa. Here's where much of the activity, with respect to mergers and affiliations, has taken place over the last few months. Take, for example, Sprint and MCIWorldcom's $115 billion announcement back in September.

Until the start of this past November, there were two realistic broadband options: cable and DSL. A recent FCC ruling, which allows satellite-transmission stations to carry network channels, might help that medium in rural Indiana. There, they don't care if the local news happens to come from Chicago; they're just happy to get coverage at all.

While satellite transmissions boast a great picture, they do suffer from a slow download rate for Internet access (in the range of 300 kbits/s) and nonexistent upload (a phone line must be used). These factors make it highly unlikely that satellite will be the medium of choice in more populated areas. Cable and one or other various DSL versions will dominate. Integrated-services digital networks (ISDNs) will go away, while T1/E1 lines will remain popular for the installed base, comprising mainly offices.

While a head-to-head technology comparison of DSL versus cable might explain why one is better than the other, their relative potential for success in this new era has more to do with business and coverage issues than anything else. Yes, DSL is easy to install. The splitterless G.lite version even obviates the need for a service technician to come to your door. On the other hand, cable is already installed in a vast number of households.

The Next Decade's Workhorses
That's actually part of cable's problem, however. The cable companies didn't tell us that our service degrades as our neighbors come online because they also see the advantage of an already-installed medium. And both technologies are plagued by coverage and quality-of-service issues. They may soon be surpassed by a rising level of support for alternative media, such as fixed wireless and fiber-optic lines—the workhorses of the next decade.

Thankfully, there's been substantial research into a technology called vectored orthogonal frequency-division-multiplexing (VOFDM). A fixed, wireless local-loop connection might soon be realized in a multichannel, multipoint distribution service (MMDS) scheme. This holds the promise of local, high-speed, indoor wireless connections in non-line-of-sight environments.

VOFDM is key because it allows multiple channels to lie side by side, with minimum bandwidth requirements and up to two usable antennas. These antennas can detect and, hence, decode the transmitted signal accurately, despite multipath interferers. The wireless connection may enable a typical data rate of 20 Mbits/s in both directions in a 6-MHz channel. Doubling the channel to 12 MHz could give a throughput of up to 40 Mbits/s.

Fiber-optic connections to the home remain the ideal, but they're still a long way off. Going wireless, as with VOFDM, has the advantage of wide coverage (up to 10 miles) and a minimum of hardware investment in terms of laying down lines or purchasing dark fiber. But like most technologies, its success depends heavily on how quickly it can be disseminated.

Beyond the realm of the house or office, things get a little trickier. Mobile reception and transmission have been available to the masses for a number of years. For the most part, they're even quite adequate. But that's all changed with the desire for broadband access. The 14 kbits/s available on the typical wireless phone is laughable in the face of a burgeoning need for the 150-kbit/s, 384-kbit/s, or even 2-Mbit/s data rates being planned for third-generation (3G) wireless phones.

It's a straightforward matter to design a wireless, Internet-enabled telephone capable of 114-kbit/s reception. In practice, though, you'll need a lot of luck finding support for that rate. The norm is 14 kbits/s. This problem has to do with coverage and infrastructure. Billions of dollars are invested in the current network of cellular base stations. To upgrade would incur losses on the carrier's part.

They can, however, configure the base stations for upgraded service at 150 to 384 kbits/s. That's most likely to happen over the next year or two, at the expense of user density.

This move will up the data rate sufficiently until the standards for 3G devices are set. By 2002, these devices should begin to roll out with global roaming meeting the 2-Mbit/s rate guidelines set by the Universal Mobile Telecommunications Service (UMTS).

Portable, handheld devices are swooping down to take advantage of the upcoming megabit data rates. But what technology is going to display the exhilarating graphics that such rates are supposed to enable? Monochrome, passive-matrix, direct-view displays have their place on the latest web-enabled devices that connect to specially designed versions of popular web sites. They're particularly attractive from a cost and power-consumption point of view, while being very suited to the slow data rates of low-information-content images. But these displays will quickly lose their luster when color and higher-resolution images become the norm.

As an alternative, the large active-matrix LCDs that are over 4 in. in diagonal are out because they consume too much power. But help is on the way. Research into organic LEDs by a number of companies is finally paying dividends. Organic materials are readily patterned onto large surfaces and tend to hold their structural integrity better under stress, which makes them amenable to large displays. Unfortunately, success to date has been limited to the primary colors of red and green. Blue demands a higher voltage, which isn't what a portable device needs to be generating.

Electronic Ink For Low Power
Another interesting choice, albeit for monochrome, text-oriented displays or electronic newspapers, is "electronic ink." Essentially, it's a black ink in a clear plastic shell that gets deposited onto a surface. The ink particles are oriented for dark or clear positions using a potential difference applied across address electrodes. This technology promises extremely low-cost, low-power operation at just 12 mW for a 12.1-in. display.

For the handheld, graphics-intensive device, however, a whole range of microdisplay options will soon be available. Some have already hit the street in various shapes and forms. The most popular and widely accepted version clips onto a pair of glasses and presents a virtual display at high resolution (1024 by 768 pixels) with 16.7 million colors.

Using field-sequential color filtering, the displays have a refresh rate well beyond the 72 Hz required by the Video Electronics Standards Association (VESA). They're sure to be a very popular solution over the coming years as graphics-intensive Internet portals go mobile. In a nutshell, expect poor graphics for now. But by late this year, high-information-content microdisplays will be well within the price range of most designs. These will be followed by organic LED displays by mid-2001. Electronic ink will find its niche in the portable, foldable, electronic newspaper-type device.

Over the next couple of years, then, portable devices and high-speed office and home connections will become widespread, dishing and receiving information to and from the backbone at phenomenal rates. This will put pressure on the links, or ribs, that have already grown to form that connection. New, faster links will be necessary. Metropolitan optical networks will rise with media-independent access gateways that will take DSL, cable, POTS, or wireless input and shovel it to the backbone at reliable, terabit-per-second rates (Fig. 2). The protocol or media will cease to be a bragging point. Look for the appearance of ICs that can handle multiple protocols, as well as optical switches that are now emerging from research labs.

These switches are mostly based on microelectromechanical systems (MEMS) technology, in which microscopic mechanical mirrors route optical signals at terabit rates (Fig. 3). The ultimate goals are all-optical computers and networks that eliminate the need to constantly transition between optical and electrical signals for amplification, regeneration, and routing.

Until now, the LAN and wide-area network (WAN) have been almost completely separate entities with their own protocols. Typically, it was Ethernet for the LAN and asynchronous transfer mode (ATM) for the WAN. This setup may soon change, however. The popularity of Ethernet in the LAN has been such that a 10-Gbit/s version is being worked on right now. It's likely to get somewhat of an identity crisis, because such data rates, when combined with its widespread acceptance and popularity, may make it a viable candidate for backbone applications.

SONET Will Lose Dominance
As the Internet continues its meteoric rise, the need for "packet-over-everything" may take its toll on the synchronous optical network (SONET), which has been the workhorse of high-speed data-transfer systems for many years. Unfortunately, SONET is optimized for circuit-switched data, making it less amenable to the increasing level of packet-switched data now being transmitted. It won't be going away anytime soon, but its dominance will wane.

On the entertainment front, a lot has happened that may well accelerate the delivery of quality digital-TV signals to both our homes and portable receivers. This may be the year that digital TV finally takes off, since the release of quality 8-VSB decoders and a recent FCC ruling thwarted any attempts to rekindle the coded OFDM/8-VSB debate. To date, though, the quality of the programming still doesn't bode too well. Better make it the year 2001 before the market surpasses the "curiosity" status.

Thanks to the massive strides made in the last 20 years, we enter this millennium with a barrage of communications media. Some will come and others may go, but many will become part of the fabric that will blanket and homogenize our society. Its effects have already been felt. Uncensored communication with the outside world, for example, has been largely responsible for crashing through the walls of silence and exposing dictators and tyrants for their true selves.

On the home front, the Internet has been beneficial. But many voice concern over its tendency to isolate heavy users from true, more "natural" interaction. These arguments may soon become meaningless. High-speed, untethered connections stand to make more realistic interactions possible in places other than our basements.

TAGS: Intel
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.