The Internet of Things: Our Inevitable Future or a Pipe Dream?

Among today’s technocrats and thought leaders, few concepts are more popular and appealing than the Internet of Things (IoT). In general, it refers to a future in which networks of low-power sensors, some say a trillion of them and most of them wireless, on every “thing” from bearings to motors to patients to appliances, will provide petabytes of information. This information will let people better manage factories, airports, hospitals, homes, and their everyday lives, sometimes without them even knowing it. That’s because computers and actuators will also be on, or networked, to many of these “things” and they will take incoming sensor info and make decisions on whether to shut down or speed up, open or close, or notify managers or not—all without human intervention.

Most IoT backers envision the Web and Internet connecting all these devices. They want to leverage the standards, as well as the computing and telecommunications infrastructure already in place. But others, such as Chris Muench, CEO and founder of C-Labs, a firm that helps industrial clients prepare for the age of IoT, believes the Internet and Web are not necessary for all IoT networks. According to him, the term IoT actually refers to the Internetworking of Things. So a house or factory, or maybe an entire business, might have its own network of sensors, computers, and smart devices connected in a cluster independent of the Internet. A hub or relay might also be included and used as a portal to the Web or Internet.

So, the IoT is still evolving in concept and scope. However, like many other great ideas that promised to lower costs for everyone and simplify life—think Esperanto and the move to go metric—there are advances that support it as well as some possible speed bumps on the road to this new interconnected “smart” world.

Building blocks

Several trends are fueling the IoT wave: ever-shrinking, ever-cheaper computers and sensors; cloud computing; and wireless technology. Smaller sensors and computers need less power and space, meaning they can be put almost anywhere. This is especially true considering the growth in energy-harvesting devices that, so far, can pull a few milliwatts from waste heat, stray electromagnetic radiation, sunlight, pressure, and barely perceptible vibrations.

Wireless technology and cloud computing, helped by recognized standards, make it easier to put sensors and computers, as well as software and data collection and processing, in remote or inaccessible locations by eliminating the need to run wires to them or have co-located hardware. This makes IoT attractive to companies running vast networks, such as power companies; freight and transportation companies on land, sea, and air; as well as franchises and financial institutes.

Industries can also make good use of wireless sensors and devices to monitor equipment for wear, vandalism, or predictable failures, thus minimizing downtime and scheduled maintenance while increasing production and throughput. It also lets them control remote and inaccessible equipment and machinery.

“The IoT will definitely impact the automation and machine design industries,” says Steve DiMarco, President of Tolomatic, a Minnesota firm that designs and manufactures electric linear actuators, pneumatic actuators, and power-transmission products. “Costs will decline due to the mass production of communication components that enable networked, digital communications in places where it hasn’t been cost effective in the past. Sensors, actuators, and other devices will communicate status and diagnostic information that will let machines run more efficiently and be serviced more effectively. And this use of real-time operating information will expand exponentially as the cost of connectivity comes down.”

To prepare for the IoT, Tolomatic has been embedding Ethernet communications in controllers and drives—a key step on the road to networked data sharing.

Stumbling blocks

A major set of stumbling blocks to IoT is the lack of standards unifying all the equipment and software that connect the things. However, one standard’s hurdle—the number of available Internet Protocol (IP) addresses—was at least partially overcome.

Every device on the Internet needs an IP address that defines its location and identity. Until recently, devices used IPv4, a 32-bit address first described in 1981. It provides 2³² unique addresses, or about 4.3 billion.

A new protocol, IPv6, was described back in 1998 and it’s a 128-bit address. This means it has 2¹²⁸, or 3.4  10³⁸ (or 340 undecillion) addresses. That should be plenty of addresses, but so far, only 4% of the world’s devices are estimated to be using IPv6. And IPv4 and v6 protocols are not compatible. Fortunately, there are fixes that let IPv4 and v6 hosts communicate. Still, if 96% of all devices on the Internet will need new addresses, that could slow the IoT wave.

But other standards are lacking, such as a uniform way to identify what a device is and can do, what it’s connected to, and how to communicate with it. There are companies trying to accomplish this task with databases and registries.

The Wireless Registry, for example, lets users claim addresses and identify their wireless devices. Then, when combining that with an application programming interface (API) from the company, those devices can identify other (registered) devices around them without connecting to them. Although the first device registered is free, it cost about $5 annually per device for the rest, a potential upside of billions per year for the Wireless Registry. But it’s highly likely it will face a host of competitors, some with very deep pockets.

It’s not like the early days of the Internet, when standards and protocols were developed by individuals and institutions trying to develop and define a truly universal and open system. These days, companies play hardball with standards, trying to create a playing field tilted totally in their favor. Ease-of-use and universal compatibility—forward, backward, and with other devices and networks—might be preferred by users, but not to those trying to maximize profits by selling hardware, software, and services.

Another major stumbling block is security. Software networks apparently aren’t that hard to vandalize. Ask Sony, Target, or the NSA. And they could afford IT security firms to harden their networks against hackers and itinerant bugs and viruses.

Older devices are particularly weak spots in terms of security. What was once a software feature could become a portal for hackers. For example, many Unix- and Linux-based computers, as well as IoT types of things like webcams, automated locks, appliances, and calculators, use Bash (Bourne again shell). Today, a bug dubbed the Shellshock Bug exploits Bash by letting hackers insert commands that get executed at the operating-system level. The National Institute of Standards and Technology rates this one a 10 on a scale of 1 to 10 (10 being the worst).

While patches and updates exist that keep the Shellshock Bug at bay, most consumers will not know which of their devices use Bash or be aware of the fixes. They will be vulnerable and not even know it.

The IoT adds another layer of complexity that’s ripe for hackers: The ability to control devices, not just download or corrupt data files. Thus, companies might be hesitant to put the control of planes, trains, hospital equipment, and other devices on the IoT. And consumers might not jump at the chance of putting control of home monitors, webcams, locks, and automobiles at the mercy of the Internet.

The emergence of two IoTs

Given the stumbling blocks and potential payback to users, it’s likely two types of IoTs will emerge: one for consumers and another for industry. The industry version will probably appear first—in fact, it’s already here in large part.

Proprietary networks for collecting customer data linked to machines that write and send out bills are common. So, too, are high-speed links for sensor data and actuators in nuclear plants and power grids.

The mining industry, for example, has long used satellite communications to monitor wells, according to Greg Montrose, marketing manager at American Sensor Technologies, a maker of LVDT, pressure sensors, and other transducers. “But with the increased infrastructure of cell towers and decreased cost of wireless devices and data packages, it is much easier today to communicate information for sensors in remote areas via the Internet.”

Industry holds a major advantage over consumers in having staff on the payroll to handle compatibility and standard issues. They ensure sensors, actuators, computers, and networks will work with one another, as well as with legacy systems. In addition, funds are set aside for regular hardware and software upgrades, all tax-deductible. IT staffs also stay up-to-date on software bugs and viruses, as well as the fixes, firewalls, and defenses against them. Everyday consumers jumping on the IoT bandwagon are left to fend for themselves and many lack the inside knowledge to manage data networks.

Consumers might also have a hard time quantifying the benefits IoT brings to their lives. Being able to answer your front door on a smartphone when you’re away or having your furnace reduce its output because the dryer is running don’t generate ROIs that are easily computed, if at all. Instead, consumers will have to want the services at the prices they can get using the IoT.

But in industry, accountants will closely track IoT expenditures and benefits. The benefits could include better quality, lower energy use, quicker times to market, better use of personnel and production equipment, closer ties to customers and suppliers, and, most important, higher profits. Meanwhile, if widespread use and adoption of IoT technology makes sense in any of these areas, companies will invest the time and money to deploy sensor and data networks.

Industry also has another advantage. It can bankroll R&D into analytics, discovering ways to process large flows of sensor data into a competitive edge. The IoT gives companies access to large amounts of data (aka Big Data), and many of them are well positioned to take advantage of it. Consumers aren’t as well positioned, nor do they have large amounts of money to invest.