Interview Prep: M2M Magazine

January 20, 2007

Interview Responses, M2M Magazine

Note: This piece was written in response to a request by M2M magazine for a formal interview with the CEO of Microwave Data Systems. The questions were provided by the magazine. I prepared these answers in extended written format for the magazine, and to be used as reference material, if needed, prior to and during the interview itself. The style is informal - almost conversational, in keeping with the nature of the request.

M2M Pioneers is the cover story of the March/April issue of M2M Magazine, which covers machine-to-machine telemetry and remote monitoring. The article will profile 10 individuals with the greatest influence and vision in the industry; they are the people most responsible for shaping where the market is today and where it is headed.



A radio - any radio - is fundamentally a signal processing device. It transforms electrical signals into radio frequency signals, and vice versa. The ability of a radio to do this well is of course limited by physics, which dictates the characteristics of signal propagation and degeneration over time and distance, according to the frequency or frequencies in use. In traditional analog radios, that pretty much sums up the challenge; anyone familiar with AM and FM radios will understand many of the limitations imposed by the physical universe in areas of sensitivity, interference, distance from the transmitter, signal quality, etc.

It turns out that digitizing an analog signal allows a clever designer to begin to overcome the limitations imposed on analog signals. More information can be carried within a given frequency range. Better sensitivity can be produced. Improved tolerance for interfering signals can be achieved. Higher rates of information transfer can be developed. Lost signals can be reconstructed.

The advantages of digitizing analog signals are now widely recognized in the wireless industry, as they are in audio, photography, and more recently, video production. It’s easy to forget, though, that this was not always the case. Digitizing the signal properly was difficult, processor-intensive, and introduced other challenges, such as latency increases. The arrival of silicon-based DSP capability helped immensely; DSP chips provided extremely high-speed mathematical operations, such as matrix multiplication, that were necessary to approach the real-time signal processing demands of high-speed digital wireless communications.

MDS was the first to master these challenges for industrial wireless data communications and deliver digital solutions to the markets we serve. The result was wireless transmission characteristics that bested their analog predecessors by a wide margin. The MDS iNET 900™, for example, was the first industrial Ethernet solution available; a long range, high speed, industrial wireless IP/Ethernet solution, it allowed customers to bring mission-critical, revenue-generating data from fixed assets over Ethernet or a serial gateway and onto digital networks.

MDS initially chose to focus on the utilities/energy sector because the company’s founder identified a specific customer need in this market that was not being satisfied. It was a marvelous confluence of technical capability with industry knowledge and customer need - a philosophy and core competence that MDS carries beyond the utilities sector into other industrial markets such as oil and gas, public safety, rail, and traffic.


This is an interesting question, because, in some ways, traditional utility networks and the Internet technology have characteristics that are diametrically opposed. Utility networks, for example, are of necessity highly deterministic. Do this first, then do that, carefully, in order, except under these circumstances - double check, triple check, no mistakes -- it’s all very orderly and painstaking, because it must be. Pumps, valves, switches - these must be controlled and monitored with precision and careful timing, because the consequences of failure can be dramatic.

Internet protocols, by contrast, are non-deterministic by design. IP networks accommodate bursty traffic extremely well, but are less effective at quality of service guarantees. TCP is reasonably efficient, but makes assumptions during failure conditions that are not well suited to the inherent unpredictability of a wireless transmission environment.

The challenge is to exploit the well-developed and extremely popular standards of the Internet domain, but implement them in such a way as to protect and extend the strengths of traditional utility networks. The key is to extend, rather than replace, existing network infrastructure - something we specialize in at MDS.

Today, our typical utility/energy customer is seeking to extend the corporate network to the field. This means offering desktop capability remotely, sometimes at long distances, and increasingly on the move. Ideally, fixed assets and mobile data users can share the same physical infrastructure. We meet this challenge by offering the longest range communications solutions, which enables near real time access to data, fixed or mobile, using standards based interfaces.


Many years ago, Sun Microsystems came up with the clever tag line “The Network is the Computer.” It wasn’t really true then, and it isn’t really true now, but it does offer some insight into a larger debate about where intelligence in the network should properly reside. Before the personal computer ushered in the widespread democratization of computing resources, “intelligence” was almost exclusively restricted to the back office machine room. Then, these same PC manufacturers began to improve intelligence at the edge of the network, and telecommunications providers began building intelligence into the network itself, with services like speech recognition systems, centrex, and packet switching infrastructure.

The debate, as it turned out, was not properly about where the intelligence would or should reside - it was developing everywhere - but rather, how the intelligence throughout the network could be made to interoperate. The gating items became standards development and legacy systems support - not technology - and this proved to be true in software, hardware, networking, and wireless communications.

About five years ago, after Internet-based communications standards established critical mass, data networks actually eclipsed traditional circuit-switched voice networks in popularity, and today we see the convergence of all kinds of previously disparate networks under the rubric of the Internet, including voice and data, wired and wireless, fixed and mobile, public and private, long-rage and short-rage, industrial and commercial.

MDS’s role in this massive convergence is threefold, and occurs at the intersection of computer networking and wireless data transmission.

First, we strive to accommodate the widest possible variety of relevant computer networking standards in our products, in order to accommodate large legacy infrastructure while simultaneously offering a seamless upgrade path for modern technologies.

Second, we strive to transmit that data wirelessly over the longest possible distance, at the greatest possible speed, within existing regulatory constraints, and while employing leading-edge security capabilities.

Third, we do this in an “industrial” capacity, which means facilitating outstanding reliability and scalability, and the ability to operate even in extreme environments with MTBF rates measured in decades, rather than in years.


The ultimate value of any machine is how well it serves humanity. Machines, regardless of their sophistication, are ultimately tools that are intended to ease the burden that would otherwise be borne by human beings. At a very basic level, when one machine communicates with another, it relieves a human being of the need to do so manually, and that is all to the good. M2M, then, is ultimately about automation, and that is a business that we know quite well at MDS. The market is hardly “new,” but new opportunities arise constantly as technology accelerates to meet advancing industrial needs.

In our markets, automation is valuable for a number of reasons. Often, sites being monitored are far distant from the home office; manual intervention is therefore extremely time-consuming and commensurably expensive. Frequently, the machine in question operates in an environment that is physically dangerous to a human being. Perhaps most important, however, is the opportunity cost; aberrant behavior by one machine can lead to disastrous consequences if not captured, analyzed, and corrected in real time.

MDS products can be thought of as an enabling technology for rapid, reliable, highly-secure machine-to-machine communications over long distances and in extreme environments. We’ve been providing wireless solutions in the M2M space for 21 years, and our industrial wireless networks can be found in over 120 countries. It’s a track record that speaks not only to the quality of our products, but to the undeniable need for, and value of, M2M communications as the foundation of an increasingly intelligent and widespread networking infrastructure for our society.


Our customers will experience changing requirements in three primary areas: practical growth, communications heterogeneity, and network management.


By design, industrial infrastructure has an extremely long life span. It is expensive to install, expensive to remove from service - even temporarily - and expensive to replace. Most of our customers, further, have a massive installed base of legacy infrastructure which must be respected and accommodated. As such, it is both a strength and a bottleneck for the deployment of new technologies. One important, well-established, and ongoing need for our customers, therefore, is the ability to work with and extend existing infrastructure without requiring its wholesale replacement.

With that in mind, it is nevertheless clear that as customer demands grow, infrastructure must grow to meet the need. Growth occurs in two primary domains: capability, and size.

Capability growth implies improvements in site diversity, process redundancy, the proliferation of ever more intelligent PLCs, EFMs, RTUs, and the expansion of fixed assets to include mobile points of presence.

Size growth is more prosaic, but no less important; more remote monitoring points, larger volumes of data, increases in the demand for real-time responsiveness, and the accommodation of sites that might once have been regarded as too inhospitable for any practical application are all in the mix.

The key concept in understanding how these areas of growth must be approached is practicality. Varying degrees of in-house expertise, human and capital resources, and budgetary considerations all dictate the development of hybrid networks that can accommodate both aging infrastructure and modern technologies as they become commercially viable. Backward compatibility and interoperability become paramount. It is first and foremost a practical response to the legacy challenge.


Beyond growth, it is clear that the types of communications that our customers require are expanding and changing, and this change is precipitated by the raw capabilities of new technology. Alarming mechanisms, for example, are evolving from audiovisual indicators through slow-scan video capture to full-motion video. Telemetry data, once limited by slow baud rates, is becoming voluminous as advances in speed offer “fatter” transmission pipes. Remote reception points, once exclusively fixed, are becoming nomadic and even fully mobile. That change alone offers an entirely new world of applications possibilities for workforce automation solutions that were barely contemplated even a decade ago.

Network Management

The final challenge, as networks grow in size, complexity, and heterogeneity, is network management. Utilities must be able to monitor and manage their networks centrally, using a common network management platform that provides detailed real-time access to every fixed and mobile point in the network, as well as the communications links that connect them. It is important to recognize, however, that not all network management capabilities are, or should be, central. Distributed network management using standards-based tools and protocols is increasingly vital for a flexible, responsive networking infrastructure.


As I mentioned before, we are inevitably constrained by the laws of physics. Wave propagation in the physical universe happens according to very specific and well-understood laws that govern the interaction of power output, power consumption, data speed, transmission range, sensitivity, interference robustness, line of sight characteristics, etc. It’s a complex mix of sometimes competing characteristics. MDS has always striven to offer an optimal balance of range and speed, without sacrificing too much in other areas of concern.

Here’s an example. Many of our radios are designed to run exclusively on solar power. This is often a requirement for unmanned operation in a remote site where there is may be no other source of electricity. This requirement inevitably means that the radio must consume only very minimal power (for obvious reasons). That means that the amplifier in the radio has less power to work with, which inevitably constrains the dynamic range (the ability to listen to strong signals and weak signals simultaneously) of the receiver. It’s physics. Clever design helps, of course, but it’s an example of one of many tradeoffs that RF designers must grapple with daily. MDS strives to engineer radios that (in this case) find an optimum balance between power consumption and dynamic range, without emphasizing one at the expense of the other. Anybody can do one or the other - but to do both well is a difficult engineering challenge, and is precisely the approach at which we excel.

I spoke earlier about the ‘intelligence’ of the network, and where it might reside. Industrial radios are traditionally thought of as ‘transparent’ devices; that is they transmit data without knowing, or needing to know, anything in particular about the data being transmitted. But radio frequencies are a “rare” and increasingly expensive resource - particularly in the licensed frequency ranges, and the challenge is to maximize the efficiency of those RF resources by minimizing waste, overhead, and error. Adding intelligence to the radio itself - knowledge of the protocol in use, for example, permits us to improve transmission efficiency and thereby increase the effective bandwidth of the signal without increasing the actual bandwidth.

Moving forward, then, it’s clear that our pipes will continue to get fatter (more raw data transmission capability) and more intelligent (greater efficiency). Coupled with our well-established engineering capabilities, and our philosophy of offering an optimal balance of the many characteristics of RF transmission, we are confident that MDS will continue to enjoy a leadership position in the markets we serve.



Unsurprisingly, perhaps, the most important overall trend across the markets we serve, and across the technology sector in general, is the transition to wireless communications. This is happening at an astounding pace, and there is no end in sight. Public networks, such as the telephone network and the Internet, are exhibiting a rapid transformation from wireline to wireless communications in the last mile, and the capacity and capability of last-mile access is trending with Moore’s law, in which price/performance doubles about every 18 months. Commercial wireless communications are now a fact of life in much of the developed world, and public expectations for high-speed wireless connectivity are now part and parcel of lowest common denominator discussions for infrastructure improvements in the public sector.

This is an important trend not because it affects markets that we serve directly, but because of the corrollary effects that public networks have on our core markets in psychological, financial and security domains.

Psychological Impact

The public Internet is rapidly becoming an indispensable fact of daily life in North America, and it is only a matter time before municipal, state, and national governments step in to regulate it as a public utility. This is already happening - cities such as Philadelphia are providing widespread wireless ‘hot spot’ network connectivity using WiFi technologies, for example, and public expectations for ready wireless access to the Internet are growing in tandem.

This is an extremely positive development, but it carries with it some important challenges. First and foremost, no network, wireless or otherwise, can be affordably engineered to sustain maximum demand at all times. In practical terms, this means that networks tend to fail during periods of emergency - not only from physical damage, but from saturation. That means that mission-critical networks, such as traditional utilities, cannot afford to rely on public network infrastructure for their day-to-day operations; the risks are simply too high. The best alternative is to construct a private wireless network for support of these mission critical activities, and, happily, private networks are exactly what MDS builds. You can expect, therefore, to see private wireless networks to grow in both number and size as public wireless networks expand.

Financial Impact

Wireless networks have become far less expensive that their wireline counterparts. This is true in both procurement and operation, and it is particularly true for networks that cover large distances or exceptionally difficult terrain. Developing countries around the world are now simply bypassing wired infrastructure altogether. Wireless options provide similar or identical performance and functionality at far lower installed cost, and with a substantially greater degree of overall reliability.

The other side that coin is equally valid in our markets - the growth of very small-scale wireless networks with limited range. These types of networks essentially perform two functions. First, they simply replace wires. Think Bluetooth or ZigBee for short-range wireless I/O, for example. Second, they provide new functionality, such as wirelessly connecting a police officer’s PDA with his radio gear in the police car, or providing VOIP data transmission to a firefighter inside a burning building.

These capabilities have be available for some time, but they have only recently become sufficiently affordable for widespread adoption. The bottom line: wireless works better, and costs less.


No wireless conversation is ever complete without a consideration of security. Today, even the utility industry, which is traditionally slow to adopt leading edge technologies, is now keenly aware of the risks associated with insecure operations. Wireless networks today employ a variety of layered security technologies, including sophisticated AES encryption, RADIUS authentication, and Spread Spectrum Frequency Hopping. For all practical purposes, nefarious activities such as eavesdropping and unauthorized network access are virtually impossible. In the final analysis, wireless data transmission can be made at least as secure, if not substantially more so, than its wireline counterpart.


Beyond the wireless transition I’ve discussed, the most significant market force in evidence is convergence. Network convergence is all about interoperability. It means the whole is greater than the sum of the parts. More significantly, however, it means that the parts can actually talk to one another. Convergence builds upon open standards to product a logically cohesive network out of physically disparate components. Convergence is in its infancy in certain markets, and mature in others, but it is developing across the board as the foundation for the wireless networks of the future - a future in which MDS hopes to continue to play a meaningful and important role.

January February March April May June July August September (1) October November December (2)
January February March (2) April May June July August September October November (1) December
January February March April May June July August September October November (1) December
January February March April May June July (4) August September October November December
January February March April May June July August September October November December
January February (1) March April May (1) June July August September October November December
January (3) February March April May June July August (1) September October November December
January February March April May (1) June July August September October November December
January February March April May June (1) July August September October November December
January February March (2) April (1) May June July August (1) September (1) October (1) November December (1)
January (6) February March April May June July August September October November (2) December (1)
January February (1) March April May June July (1) August September October November (1) December
January February March April May June July (1) August September October November December
January February (1) March April May June July (1) August (1) September October November (1) December
January February (7) March April May June July August September October November December
January February March April May June July August September October November (1) December
January February March April May June July August September October November December