What is the Internet of Things (IoT) and what makes it different from the traditional Internet?
This is important to architects because… IoT is a key emerging set of technologies that logically extend the technologies that architects are already working with.
MULTI-PART BLOG POST – In part II, we will cover architecture, design, and security.
Scope of Discussion. We assume that readers understand that IoT can be applied to various domains such as home devices, medicine, agriculture, and manufacturing. When we want to cite a concrete example, we’ll use the example of an Internet-enabled soda machine.
Modern IoT solves the “Basket of Remotes” problem. Most of us who have watched television have used a remote control. If you bought a cable television package, it has a separate remote control. Perhaps you added a DVD player, a BlueRay player, and a Video Cassette Recorder. Each one of those devices has a separate control, but they are all essentially doing the same thing – controlling entertainment media for a single TV screen. Modern IoT endeavors – successfully – to simplify analogous situations and essentially have a “single remote”.
The Internet-Enabled Soda Machine
We will give examples of problems and potential solutions based on an Internet-enabled soda machine. As you can see from the photo on the left, the Coca-Cola company has already done this.
Good example. Soda machines are good for examples. This is because most people are familiar with them and the business problems around them. Better yet, the devices are not that complicated to understand, but have many potential IoT applications.
Some Preliminary Assumptions
IoT is not new. According to Cisco Systems Dave Evans at https://www.cisco.com/c/dam/en_us/about/ac79/docs/innov/IoT_IBSG_0411FINAL.pdf, there were already twice as many devices on the Internet as people as of 2011. In 2019, there were over eight billion devices connected to the Internet. The trend is now accelerating. There are some great statistics available at https://iot-analytics.com/state-of-the-iot-update-q1-q2-2018-number-of-iot-devices-now-7b/
A familiar example. The NEST thermostat pictured to the right is a good example of an IoT device. It is a good example because it is self-contained and specifically designed with communications, learning, and analytics in mind.
The definition of IoT is evolving. The definition of IoT of five or six years ago has changed. IoT is just a blanket term for the technologies used to extend the Internet to machines. Specifically, “machines” do not include traditional devices such as desktops, laptops, and smartphones that are essentially general-purpose computers that are attended by human beings. Thermostats and soda machines are great examples of such non-traditional Internet hosts.
The devices are not the key enablers. The key enabling technologies for IoT actually have little to do with the devices themselves, but rather adapting existing Internet technologies, analytics, and machine learning to arrays of devices. Analytics and machine learning take on increased importance with IoT because traditional Internet hosts interact with humans, in which case the human is providing the analytics and the learning. The devices are the most visible part of an IoT system, but not the most important.
KEY POINT: Do not focus on the devices. The key focus is on the information system infrastructure for implementing smart, connected objects.
Clarifying some confusing terminology. IoT concepts have been around for a long time, in fact predating smartphones, tablets, and even the Internet itself. Terms such as “supply chain”, “smart objects”, “industrial internet”, “machine to machine”, “avatar”, “pervasive computing”, and “ubiquitous computing” can all be understood as referring to IoT. Marketers and salespeople also tend to invent new terminologies for the same thing, thus adding to the confusion.
KEY POINT: While terminology can be unclear, the common thread is application of existing Internet technologies to arrays of devices. If we are applying existing Internet technologies to arrays of devices, that is IoT.
Subdivisions of IoT. The key subdivisions of IoT are typically cited as residential, commercial, industrial, and infrastructure, each of which have broadly different technical and business assumptions. Our IoT-enabled soft drink machine is a commercial IoT application.
IoT and Technology Convergence
IoT is of increasing importance now because of the “convergence” of various technologies. There are several independently developed technologies that, when combined, enable a robust IoT infrastructure.
The diagram below above is in layman’s term. Some of the technologies above reflect electrical and mechanical engineering advances. Experienced software developers can, however, map many computer technologies that have emerged over the last several decades onto the timeline above:
(1) Geolocation. GPS technology is used to locate things. Example: By knowing where our soft drink machines and supplies are, we can manage stock levels, distribution, and manufacturing more effectively.
(2) Cost Reduction. The cost of computers, electronics, and software systems continues to reduce. Example: If it costs $1000 to monitor a soft drink machine, we probably won’t do it. If it it costs $5 to monitor a soft drink machine, however, we probably will want to do it.
(3) Miniaturization and power-efficient electronics. Most developers are accustomed to thinking about electronics in terms of computers. There are all kinds of electronic devices, however, that are not related to computers. Almost all electronic devices cost less and consume less over time. Consider the simple examples of a camera. In 1954, the RCA TK-41C on right weighed 310 pounds and was mounted on a dolly.
Compare that with the 2018 Adafruit TTL Serial camera to the left that is one inch wide, costs $40, and is more effective.
This type of miniaturization applies to all types of sensors, to include light, temperature, touch, force, flow, chemical, pressure, magnetic field, weather, humidity, moisture, vibration, sound, and electric field sensors.
Example: Our soda machine can make use of almost all of the sensor types above. For example, a temperature sensor can indicate that a door is open or that maintenance is needed.
(4) Teleoperation and Telepresence. An experienced developer knows that the emergence of cost-effective worldwide communications, ubiquitous wireless networks, and the ability to run low-cost servers on a device to receive instructions allow for teleoperation and telepresence. Example: An operator on the other side of the world can check the stock in the soda machine with a camera to determine if more stock should be ordered.
(5) Advanced Sensor Fusion. Advanced sensor fusion – groups of sensors communicating and working together – implies that analytics technologies are being applied. Example: Instead of a human checking the stock through a camera, a camera can communicate optical characteristics and relative machine positions. When combined with pressure sensors and the external temperature near the machine, stock can be ordered automatically by a computer rather than a human.
(6) Software Agents. The presence of software agents implies that machine learning or other artificial intelligence technologies are being applied. Two key areas that are changing and accelerating IoT are “ambient intelligence” and “autonomous control”.
(a) Ambient intelligence. This refers to the capability of devices to accommodate human beings. The machines would recognize you and your situational context, tailor themselves to your needs, change in response to you and anticipate your desires. Example: My soft drink machine knows that I often go home late and don’t want caffeine in the early evening and thus might suggest a fruit-flavored beverage on the way out of the office.
(b) Autonomous Control. In this case, a machine can make a decision on it’s own. Example: No one has gotten a soft drink for hours, so the unit decides to turn off the refrigeration for the day and start it again at 4 AM in time for the next day.
(1) Definition of IoT. The broad definition of IoT simply implies extending traditional Internet technologies to machines as well as people.
(2) IoT becoming more important. IoT has actually been around for 20 years, but it is more important now because the convergence of independently developed technologies allows more IoT functionality for less investment than we needed earlier.
(3) “Basket of Remotes”. The most recent developments in IoT are solving the “basket of remotes” problem.
(4) Don’t focus on devices. The devices are interesting, but most of the work for IoT is on the communications, the analytics, the artificial intelligence, and other software systems to get all of this equipment working together effectively.