FROM EDN EUROPE: What to make when you can make anything

By Graham Prophet -- EDN, 12/7/2000

At November's Electronica show in Munich, Germany, I attended a presentation by Agilent. The presentation featured, among other topics, an instrument (a 26.5-GHz spectrum analyser, on which you'll find more information elsewhere in this issue) that seems to sum up many of the trends that our ever-more-digital industry constantly wrestles with. Simply, when you look at a block diagram of the instrument, there is little that a designer from the Hewlett-Packard of, say, 25 years ago would recognise as part of a spectrum analyser at all. The instrument has an RF front end that is fairly familiar (at least in principle) and some mixing and attenuating circuit blocks, but that's about it. Then, you have an A/D converter, and the rest of the instrument is basically a RISC-based workstation in an instrument case. With the exception of those front-end signal-capture functions, all of the defining characteristics of the instrument are attributes of the software.

To illustrate the application of today's instrumentation in the creation of tomorrow's products, Agilent has produced a video charting the development of an imaginary future product. The video is, to say the least, fanciful, but it's amusing enough in its own way. The envisaged product is a portable cardiac monitor for someone unfortunate enough to already have health problems in the cardiac department to carry. It boasts all the trappings that tomorrow's communications technology can bestow. It has (we infer from the features depicted in the story line) full 3G cellular communications with flawless video. It has an integrated GPS to tell the ambulance crew where to find the patient, and it wirelessly connects (Bluetooth, presumably?) to the patient's implanted cardiac monitor and to the implanted and autonomous defibrillator. (A somewhat alarming concept, you might think.)

Of course, the hapless patient in the story is enjoying his day in some remote countryside location when he suffers a recurrence of his condition. His personal intensive-care unit leaps into action, putting him in instant video communication with his cardiac surgeon and summoning the emergency services to his aid with a precise fix of his location. (I must admit to finding one particular moment of pure B-movie joy in the script, when, after a quick jolt from his implanted defibrillator revives the patient, he opens his eyes as the video link tells him he's "gonna be OK." But I digress.)

So, what can you take issue with in this scenario? Sure enough, nothing depicted in this product concept is impossible or is even that far away from functions that are available now. 3G cellular capability is within sight, the GPS is established, we know how to use cardiac monitors, and even the much-heralded Bluetooth is actually becoming real products. I have a concern that even those who deal with the integration problem on a daily basis—and Agilent certainly falls into this category—are inclined to make light of it when it comes to this sort of future-product crystal-ball-gazing exercise. Pulling together the disparate elements of such a product would be not just a substantial effort, but also a massive systems-integration task. And, issues such as the power budget and software verification for such a mission-critical device hardly bear thinking about.

But, never say never; pretend for a moment that it's technically feasible. What's the problem? I wonder who will be able to afford it? In the United States, some regard the cost of (mainly insurance-funded) health care as already out of control. In Europe, where there is a greater tradition of statecentric social medicine, the systems in several countries are showing the strain of coping with the costs of modern treatments and aging populations. In the terms of the video, how many cardiac physicians will there have to be if all of their patients are online and expect immediate face-to-face contact when they feel unwell?

The era of systems on silicon may bring us to the point at which the primary constraints on what we can build are no longer technological. That will not mean there are no limits; it just means that we have to examine the external constraints ever more closely. In Agilent's video, it would mean that the marketing department of the hypothetical medical-product company should be saying, "We can build this thing, but if no one can afford the medical-insurance premiums that go with the level of support it needs, what's the point?" When you can build anything you can imagine, the results of that imagination need ever-closer scrutiny.