Scarce Resource Pressures

Andy Haire
May 10, 2016


The objective of this third of four articles is to raise awareness of the influence that scarce resources play in the communications sector. For this writing, the scarcity will stay with two areas: the radio spectrum and the device identifiers. When the marketplace's government authority takes the longer view on the future of IoT devices two critical issues surface: how to communicate with and among these devices, and how can each device be uniquely addressed. Both present the designer with limitations in a market that sees unlimited opportunity.

Further consider that within the communications landscape these same resources remain in competing demand, so – no dodging the question: who deserves these resources now, and in the predictable future?

Ever since the dawn of commercial wireless communication at the end of the 19th century, someone had not only to set rules for play but also serve as an impartial arbiter. Today the industry regulator serves that purpose.

Marketplace rules provide some structure; but what forms should they take; what should they accomplish? Make no mistake, even in the face of maturing markets around us, where competition has permitted a more relaxed role for the regulator, the sector is in no position to self-regulate, given two factors: someone needs to guard the public interest, and someone needs to insure that the public’s resources are used efficiently.

Let's first turn to the basics to understand why this impartial bystander is necessary: what are the primary responsibilities when we consider radio spectrum? The responsibility is fundamental: assign who can use what, and insure when it is used that it is free from interference from others. Of course this responsibility is complex and even problematic, but to adopt a sporting metaphor: understand what the referee on the pitch wants, what will be observed, and what is acceptable play, otherwise expect a yellow card. Responsible regulators understand fully that they are guardians of the public's asset (spectrum), and they have to insure it is most efficiently used for the public's good (important to note: not necessarily the provider's good).

First area: Manage the scarce resource

This first issue considers how devices communicate; via the 'wire' or via 'wireless' – one relies on a physical attached medium and the other uses radio spectrum. If the communication is wireless, and the public's spectrum will be used, then at least three alternatives exist: use spectrum assigned to an existing licensee, such as a mobile operator, through a commercial relationship; use unlicensed spectrum, which needs no contractual relationship, but international norms must be obeyed; lastly use proprietary, fit for purpose spectrum. Paradoxically, for trans-border applications, such as computers onboard vehicles, the unlicensed spectrum holds more potential than a relationship using a mobile operator's spectrum rights.

The broad range of devices and sensors, along with the application diversity in the IoT architecture, would indicate that spectrum planning requirements are equally diverse. Some devices send small amounts of data (low bandwidth, such as residential energy monitoring or agricultural sensors), while others transmit hoards; others are time sensitive (medical or transport management) – some prefer the term "mission critical" – while others can meander (utility provider metering). We see devices that may require a small number of bits transferred most of the time, but every so often demand large streams, such as to update the device's imbedded source program. To consider that the same device holds the same spectrum needs breeds inefficiencies. Network design might demand prioritization, which in turn leads to discrimination; some justified some not. Regulators and authorities worldwide seem to be joining the network neutrality1 bandwagon in some way; but any such debate that does not factor IoT architectural needs is shortsighted, and will constrain future innovation.

There is an added problem. Given the growth of wireless platforms along with what seems to be an insatiable demand for additional spectrum globally (in the US alone, there is a projected deficit of 366 MHz of spectrum by 20192 in the US marketplace) where will this additional spectrum come from? By extension IoT applications and platforms requiring wireless access will compete for the very same resources as other platforms such as, say, emergency services.

Raised earlier, there are three possible paths to determine where spectrum can be deployed. Unlicensed spectrum band, sometimes called the ISM3 band, is often referred to as 'license exempt', and is used popularly for cordless phones, children's toys, Bluetooth devices, and WiFi communications, etc. Permission, as a global practice, is not needed from a regulator, just compliance to worldwide transmission specifications (at a national level power and ranges are usually defined, the FCC rules4 in the US offer an example). In turn the Authority will offer no assurance that the band will be free from harmful interference – the technical equivalent of caveat emptor (‘use at your own peril’). Depending on the application this very often is the most suitable band for IoT wireless devices. Euphemistically the band has been referred to as ‘the public park' or ‘the common' – anyone can use it as long as they observe some simple rules. So if we use the noted ISM device applications as a design guide, we know that the regulator is mostly concerned about non-interference or with orderly conduct in this unlicensed band. Their concerns are overcome by certifying the device, not the provider or the application. The same should apply for M2M or IoT devices.

What is fit for purpose spectrum? In several regulatory proceedings, the UK's in particular,5 commenters recommended repurposing existing bands assigned to older technology devices to be re-farmed and dedicated to new technology, such as IoT. Of course the downside of adopting this approach and not harmonizing this use globally is that the spectrum use becomes 'fit for purpose' – it has a narrow use in one jurisdiction.

Harmonizing technical standards remains a challenge. Regional and geographic variations surface which usually interfere with achieving widespread global agreement. Understanding the rational that the US government used almost 70 years ago to convince the world to adopt the ISM, or license exempt, policy serves as a useful example6 of harmonization in a multi-stakeholder market. At the time it was microwave ovens that might be used on ships and thus cross national boundaries. International use demanded international cooperation, which this policy eventually earned.

Using the spectrum of others is another option. Frequently this is called virtual networking, or in the wireless world, mobile virtual network operators. The commercial challenges for gaining agreement is the fact that the facilities network operator (the MNO) has less economic incentive to agree than the virtual operator, so this becomes an exercise in achieving reasonable access. The access terms deal with both price and service levels. The second issue is that the MNO's spectrum right is normally limited to the geographic boundaries of a given jurisdiction, so the usefulness of the access has distinct boundaries. Assuming that these two barriers can be overcome, then the spectrum bands usually offered are 'clean', meaning free from annoying interference.

Second area: Manage identification plans

The second of these issues, device identification, should not escape attention – how do you find or address a device within a particular application or platform, especially if it is within a public network such as the internet or part of it is within closed networks? Some will argue the traditional IP addresses (especially once IPv6 is broadly deployed) are a satisfactory start with sufficient governance mechanisms in place. Others will argue that traditional 'telephone numbers', sometimes referred to by industry as E.1647 numbers, should work when needed. Other suggestions include using E.212 IMSI8 – mostly used as country identifiers in the global mobile network; and proprietary address schemes.

All of these are finite sets of numbers, and thus are scarce, once the architecture embraces one convention, it generally precludes using a different one. Scarcity will always be debated; one person's constraint is another's unlimited supply. On a personal note, the first computer I designed when I began my working career at IBM, arrived with an 'unlimited' amount of external storage – 7.5 Megabytes; and there were two disk drives; I couldn’t imagine ever using every bit of that! Another factor is that legacy numbering schemes are under the control of individual country level administrations, and there is little harmonization once the initial parts of the E.164 recommendations are met.

Systems architects understand that such hybrid design conventions will lead to future constraints. The core concerns remain: IoT devices not tied to geography should not use an identification scheme that is. Portability or mobility or nomadic capability is no longer part of market opportunity. The mobile sector overcame some of these limitations with the concept of a SIM card and even the IMEI (an identifier in each handset). With that a new set of regulatory challenges arrived; one being mobile roaming retail charges. Oddly this became a consideration for IoT regulatory concerns – and will be covered below.

What is on the regulator's mind, and why?

What are the policy maker’s concerns about such devices? Simply, gaining a level of comfort that the devices won’t disrupt an orderly market. So the regulatory authority might consider either certifying the device itself, or accept the certification from a respected impartial organization (in much the same way mobile handsets are certified today). The certification framework for IoT devices might consider: life expectancy, device ruggedness (resiliency); capability to permanently retire itself, when no longer needed; and adaptability to operate as designed without 'human' supervision. The list might expand to include security issues such as external penetrability, repurpose without expressed permission, and physical or proximate accessibility.

The regulator's nightmare: (a) unpredicted spectrum re-farming in the future, (b) noninterference from millions of low-cost devices, (c) realigning errant devices to return to performance specifications, (d) same platform for both enterprise and consumer stakeholders, (e) device to device communication – highly decentralized (security and reliability issues) – reliance on machines alone (i.e., non-human participation) to anticipate the need for recovery, (f) vastly divers bandwidth needs – latency needs – service platform needs – deployment needs, all demanding a unified air segment need.

But why give this impartial observer any control at all? In purely competitive markets the existence of competition has a natural balancing effect on efficient market behavior (both with price and conduct). This has been called the invisible hand of the marketplace by the famed economist Adam Smith.9 Economic regulators have always served as the market's proxy competitor when the real one is absent. With the introduction of device mobility, mostly caused by wireless technology, and later stimulated by data globalization caused by the internet platform, a new stakeholder arrived at the regulator's doorstep: it had to regulate someone / something it had no control over. The market also seized on this opportunity: it had a revenue stream from someone that wasn't its customer – and had no economic obligation to be efficient. Astronomical prices for roaming usage arrived, and no one could do much about it other than the consumer choosing not to consume. Consumer welfare was irrelevant and it demonstrates where authorities are powerless to play any meaningful role. By extension if the IoT device is subjected to these same high costs will it cripple opportunity for the wrong reason? The IoT developer then looks to the ISM spectrum alternative to insure the business model remains worthy, but weighs that against the potential of interference or unwanted intrusion in this unprotected band.

Will a nomadic or mobility IoT device relying on wireless connectivity and intended to work in one geographic market easily adapt to its neighboring market? Will the automobile or the tractor or the ship or the airplane be less effective once it crosses a national boundary? Will that equipment be more expensive to operate? Would that IoT or M2M provider be expected to seek permission for their devices in every country anticipated for use? That is not only unrealistic, but inefficient.


In the end, as we've just seen there are substantial gains to individuals, to economies, and to societies from IoT based applications and uses, but they arrive with a cost. The job of the policy maker is to determine where the line must be drawn between these two competing issues. A final key point: by implying there are benefits and risks makes one falsely believe that this is two dimensional – something quantitatively determined by a ledger sheet score and so having an answer. This couldn't be farther from reality; the correct balancing point is most likely a moving, multi-dimensional target that the policy will need to identify, but also be prepared to continually adjust as society's norms demand.

Let's summarize the key points made: (a) no single spectrum plan fits all IoT uses; (b) devices that are not stationary should not use identifier schemes that are also not stationary; (c) harmonize access protocols; (d) certify devices not platforms; (e) observe national data protection fundamentals, strengthen if necessary.

Finally it is crucial that public and industry conversations take place alongside the technological development. It is insufficient, arguably – naïve, to believe the IoT community can self-regulate or that the authority can develop its rules alone. Both must work together to insure a successful future for this fastest contributing part of the communications sector.



1. visited August 2015

2. Coleman Bazelon and Giulia McHenry, Substantial Licensed Spectrum Deficit (2015-2019): Updating the FCC's Mobile Data Demand Projections, The Brattle Group (June 23, 2015),


4. While somewhat dated, a useful background piece understanding one jurisdiction's ISM rules: FCC-OET; Understanding the FCC Regulations for Low-Power, Non-Licensed Transmitters, Bulletin No 63, Updated Feb 1996. Referenced March 2016.

5. OFCOM Spectrum Management Strategy, 30 April 2014,

6. Documents of the International Radio Conference (1947), Doc 1-100, P464. (

7. ITU Recommendations on E.164 (

8. Technopedia definition of IMSI (

9. University of Colorado, Boulder. (



Andrew HaireAndrew Haire with more than 30 years of experience spanning four continents has been associated with some of the industry's most successful telecom initiatives. He advises both governments and communication providers and is an expert in industry policy, market growth, strategy, technical opportunity, and economic structure. His portfolio included architecting major policy frameworks in the telecoms, technology, and postal sectors, as well as serving as regulator and ICT policy for 10 years at Singapore's iDA, soon after its inception in the year 2000.

He serves on the Board of the International Institute of Communications in London, and is Chairman of its US Chapter. Mr Haire holds a degree in engineering in the United States, and attended the advanced management program from Harvard University. He has delivered papers and speeches on policy and regulatory frameworks in Asia, Europe and North America.