5 Reasons The iPhone 6 Will Save The Internet of Things
One feature changes … everything
One exciting aspect of the iPhone 6 that has gone unnoticed is its potential to radically accelerate the growth in the Internet of Things or “IoT”. Here are five ways the iPhone 6 matters — profoundly— for this next wave of computing.
Disclosure: a company where I am a co-founder is mentioned in this post and stands to benefit from this analysis.
1. At First Glance, iPhone 6 Looks Like An IoT Non-Event
It’s been more than two weeks since Apple launched the iPhone 6 and so far, no big headlines about the IoT. Bigger screen size, better camera, increased storage, and mobile payments via Apple Pay—something many of us wish Apple had launched years ago — but the proverbial lump of coal for the IoT.
If we look ahead to the things smartphones will do for us in the future but can’t do today, the IoT figures prominently in the discussion. Construction workers, doctors, field service techs, soldiers, cops, retailers, students, and basically most human beings in the industrialized world rely on their smartphone as the device they rely on more than any other. One conclusion is inescapable: smartphones are becoming the universal remote control for the Internet of Things.
The more sensational IoT success stories of the last decade like RFID tags at Wal-Mart were built around plus-sized enterprise software, short range wireless connections, and ugly and overpriced hardware. But the consumerization of IT, low cost smartphones, and low barriers to entry for developing both smartphone apps and IoT hardware are changing the way developers think about building for the IoT. Yet despite all this progress, smartphones today are restraining the growth of the IoT by forcing developers to use wireless connections never intended as IoT technologies.
Outfitted with connectivity options that were originally invented 20–30 years ago, smartphones interact with IoT hardware nodes either via a WiFi-enabled hub or directly via Bluetooth and WiFi. There are two fundamental flaws with either approach:
- Hubs lock developers into fixed, proprietary servers. A hub is a fixed piece of hardware (like the Cisco router in your home or office) that is of little use to mobile users who need to monitor the body temperature of 10,000 head of cattle on a ranch, control a flying drone, or locate a missing Alzheimer’s patient. A hub also forces developers to interact with IoT hardware (like a home security system) via an intermediary (a hub) acting as a gateway to one or more other outdated wireless technologies like ZigBee or Insteon. As a result, developers are locked into that proprietary hub with no way of ever directly connecting to an IoT node, not unlike the days when AOL tried to convince developers that they should develop apps that access the World Wide Web via the AOL Developer Network.
- WiFi and Bluetooth are comically weak IoT connectivity options. Both were designed as streaming technologies and if they were truly compelling IoT options, one or both would already be widely deployed among the millions of commercial, industrial, or military embedded sensor networks scattered throughout the world today. But alas, this did not and never will happen. Bluetooth, after 25 years, still has woeful interoperability problems. WiFi is insecure for roaming connections, is difficult to configure, and requires a hard-coded internet server address in order to get around any of these problems—all fatal for an IoT technology. The real reason Bluetooth and WiFi reside in your smartphone is to replace headset cables and enable you to cheaply stream large amounts of data from the internet! As a result, today’s developers have to compromise with “What is the best I can do using Bluetooth or WiFi?” when they should be asking, “What sort of world-changing IoT product could I build if range, battery power, or device cost were not an issue?”
(Note: I am omitting cellular, which is expensive, power hungry, requires a monthly subscription, and is not a serious option for most IoT applications.)
For IoT unit sales to even approach analyst forecasts requires a next generation IoT connectivity option that resides in a smartphone, is not named Bluetooth nor WiFi, and can connect to IoT hardware nodes directly via a smartphone without a hub.
What does this next generation option look like? For starters:
- It should be effortless to set-up and use
- It should consume minuscule amounts of power
- It offers much longer communications range
- It is very low cost
If you agree that to ship the next 10 billion IoT devices there has to be a better way to connect IoT hardware endpoints, then keep reading.
2. Hiding In Plain Sight: Game-Changing Plumbing for the IoT
Apple’s decision to finally implement a mobile payments technology called Near-Field Communication (NFC) for both the iPhone 6 and the Apple Watch is a major event in the history of e-commerce and computing. Companies in the retail, telecom, banking, transportation, security, and gaming industries— to name a few — are all re-calibrating strategies in light of the news. Future iPads, Macs, and even Beats headphones will likely be NFC-enabled with better “pairing” between Apple products, but potentially mobile payments and wireless charging as well. Developers —and basically the entire computing industry— must now ask how future products will take advantage of NFC. With Apple’s endorsement, NFC has been transformed into the most important wireless technology in the world today and will become more ubiquitous than WiFi and Bluetooth combined.
What may not be obvious to many is that Apple added more than just “NFC” and actually added a new piece of hardware — a radio— and new radios are difficult things to squeeze into a smartphone. They cost money, suitable printed circuit board and antenna real estate is scarce, interference with other radios and antennas is common, and they can further reduce battery life. While WiFi, Bluetooth, and cellular radios have been part of the smartphone for years, NFC is the first new radio in the iPhone in seven years and this news by itself is of great significance.
Less obvious is the enormity of what Apple has (unwittingly?) enabled with NFC: a better wireless connectivity option for the IoT.
A quick explanation:
NFC is a radio frequency identification (RFID) technology, with an extremely short transmission range of 1–4 centimeters. Such short range is ideal for payments but it also means NFC is not a serious connectivity option for the IoT, which requires longer range, among other attributes.
Yet despite its very short range, NFC is an historic opportunity for the IoT: by making a tiny modification to next-generation NFC chips, smartphones are able to connect concurrently with thousands of low power IoT devices from hundreds or even thousands of meters away.
How?
NFC chips are popular and inexpensive (less than 10 cents each in high volumes) due to their use in billions of devices around the world including payment and transport cards. NFC-enabled products consume zero or minuscule amounts of power and the technology has been in use for many years. But a critical attribute of NFC as it relates to the IoT, in addition to its low cost, is the radio frequency where it operates: 13.56 megahertz. This slice of the radio spectrum is license-free and it’s already used around the world on thousands of different products.
The hidden value of NFC to the IoT lies in the ability of the current NFC antenna along with next generation of NFC silicon radios to support a second frequency, similar to the way WiFi radios support multiple frequencies for different versions of WiFi.
3. The Future Of The IoT Is (Practically) Already Inside Your Smartphone
Upgrading NFC to be “IoT-ready” is relatively simple and inexpensive:
- Re-use The Antenna. Re-using the NFC antenna is a matter of geometry that relates to electromagnetic wavelength. In other words, we can generate wireless communications with the same 13.56 MHz NFC antenna on certain frequencies, but not on others. Both frequencies must be a sufficient distance from each other in the radio spectrum that cross-interference is negligible, but close enough that the antenna will still work properly. So for a second frequency to be “eligible” it must first clear this hurdle.
- Re-use The Silicon. But an additional hurdle — re-using the same NFC radio silicon — also requires that the second frequency must also be supportable with the same semiconductor materials used in the NFC silicon radio. If the frequency is too far from 13.56 MHz, it won’t work. So in addition to antenna compatibility, silicon compatibility further limits choices of a second frequency.
- Re-Use NFC API’s. Similar to the way developers can access WiFi resources on a device regardless of the WiFi standard, IoT developers should be able to access next-gen IoT connectivity using the same or similar API’s used with NFC.
There is more than one option for a second frequency to 13.56 MHz, but the company I co-founded, Haystack Technologies, is deploying one that is well-tested: 433 MHz. Long the domain of garage door openers, tire pressure monitoring, keyless entry systems, and shipping container tracking tags, 433 MHz is an underutilized, license-free, and globally-available slice of spectrum that can deliver both long range and help deliver low power. It will not interfere with other radios in today’s smartphones — unlike radios operating around 900MHz or 2.45GHz — and its low power emissions ensure little or no health risk from electromagnetic radiation.
How much range is possible? We recently tested a beta version of our upcoming product, HayTag, at a range of more than two miles (3.2 km) in San Mateo, California using 433 MHz under strict (less than 1 miliwatt) FCC power limits and using an antenna contained in a tag that is about the size of a poker chip. Its indoor performance—the ability to maneuver through walls, HVAC ducts, metal plumbing fixtures, and even human flesh — shows great promise.
But best of all, by re-using NFC hardware and enabling a second frequency, the bill of materials cost to enable a smartphone to be truly IoT-ready is close to zero.
Still, picking a second frequency is only half the battle: we still need a 21st century data communications protocol to run on top of our newfound IoT plumbing.
4. Almost There …
All of this long range and low cost goodness requires a communications protocol that runs on top of this second frequency. For those readers who might not understand what a communications protocol is, just think of it as a way to send and receive packets of wireless data in ways that makes the most sense for the applications they support. WiFi sends packets one way to support your Skype call; Bluetooth sends packets a different way to emulate your headset cable. Same thing goes for our next-gen IoT protocol, which requires (at least) the following purpose-driven ingredients:
- NFC- and Frequency-Optimized. The protocol must be designed to work using one or more frequency bands that can co-exist with 13.56 MHz, but ideally also be designed to, like WiFi, re-use the data elements and API’s of NFC in order to minimize developer learning curves.
- Instant Connections. This protocol cannot engage in unnecessary handshaking, beaconing, Klingon language translations, or any other superfluous traits of wireline protocols like Ethernet. Connections must be made with IoT hardware in 1–2 seconds, especially with moving objects like drones, cars, or people on the move.
- P2P Discovery. Communicate directly with IoT nodes via a two-way peer-to-peer mode without requiring a cloud lookup. Nodes can “advertise” themselves or make themselves “discoverable” to other users. Think CPC ads for the internet of things.
- Ready For What Developers Bring To The Party. Our experience at Haystack has shown that with the right protocol, Android and iOS developer communities are ready to bring hundreds of new and even world-changing ideas for the IoT. The right protocol provides maximum flexibility for them almost regardless of use case, while still allowing for industrial-grade performance that supports hundreds of thousands or even millions of users. A protocol cannot be everything to everyone, but designed correctly, the right protocol can cover a vast amount of the IoT landscape.
- Ultra-low Power. Not only should the protocol not impact a smartphone battery much at all, it also needs to provide for 10+ year battery life on standalone IoT hardware nodes. I have met with hundreds of IoT developers and end users in the past 10+ years and one thing is clear: convenience is king and customers just want to set up their IoT nodes, forget about them, and let them do their job. Frequent battery recharges or replacements are the malignant tumors of many now-dead wireless solutions.
- Easy To Set Up, Easy To Use. In addition to the convenience of a low power device with long lasting batteries, the protocol must enable easy setup and connections between devices. The hair-pulling chore of pairing two previously unknown Bluetooth devices, to cite one example, is just so 1980's.
- Standardization. Like WiFi and Bluetooth, this technology will need standardization to enable cross-vendor interoperability and standardization. Ideally, the protocol will also be open source.
5. Now Ready: The Right Protocol for IoT Connectivity
When we designed the protocol stack for HayTag, DASH7, we started with the premise that the IoT needs a wireless protocol that was made specifically for the IoT, unlike protocols like Bluetooth or WiFi that were designed to pipe audio to wireless headsets or wirelessly stream Netflix!
Here are a few examples of our IoT-centricity (among many) that I believe you will be hard pressed to find occurring with other wireless IoT technologies:
Example 1: Listen Before Talk. Our hardware nodes typically stay “silent” until awakened by another authorized device. This saves huge amounts of power and is distinct from technologies like Bluetooth, which often rely on perpetual “beacons” that drain batteries and pose privacy risks by emitting a unique ID to anyone within range. Combined with our support of AES 128 private- or public-key cryptography, our approach has a superior IoT security and privacy profile to either WiFi or Bluetooth.
Example 2: Real-time Queries. In a peer-to-peer transaction between two of our devices, we enable a developer to embed a query into the very first outgoing packet sent to a second device. Why? Imagine needing to immediately know the presence or condition of a single IoT hardware node operating among 50,000 other nodes. With WiFi or Bluetooth, there is a loooong handshaking and lookup process that can clog airwaves and burn battery life, not to mention the hours it could take to interrogate 50,000 devices using those technologies. Our approach results in a response from a target IoT node in 2 seconds or less and without a cloud lookup. This is also great for connecting between one or more moving endpoints — the Internet of Moving Things— that is next to impossible for WiFi or Bluetooth.
But in addition to designing a protocol with the IoT in mind, we also designed our firmware stack for HayTag with NFC in mind, betting one day that NFC would be, you know, the hottest wireless technology on the planet. Here’s another example:
Example 3: Writing Programs for NFC is Like Writing Programs For Our Stuff. Today, an API designed to work with NFC can be extended to also work with our technology with minimal added effort as Both NFC and DASH7 use a similar language, NDEF, for accessing data stored on devices. Both also use similar foundational methods for storing raw data and database records on their chips and both have a similar interface to the “secure element” on the smartphone. Finally, there is no need the “application profiles” found in some wireless protocols: the file system and data methodology in our technology.
For companies who design, build, sell, or rely on smartphones, the opportunities provided by a truly IoT-ready smartphone are too lucrative to ignore. Use cases that were simply off the table due to the constraints of WiFi or Bluetooth are back on the table, along with the revenue and subscriber benefits that are already manifest. To name a few:
- Smart Cities
- Driverless Cars
- Smart Lost and Found
- Drones
- Mobile Advertising
- Social Discovery
- Environmental Control Systems
- Industrial Automation Systems
The importance of Apple’s NFC announcement to the future of the IoT is difficult to overstate. As an analogy, imagine the state of smartphone apps today without the ability to stream rich images or video via WiFi. The IoT is at a similar inflection point, where without better IoT connectivity in smartphones, the IoT will muddle through with underperforming hardware, underwhelming apps, a Tower of Babel of proprietary enterprise connectivity, and more of the cute-but-ultimately-optional consumer gadgetry that passes for “IoT” today.
We are excited to be bringing HayTag to market later this year and announcing other products for IoT developers in 2015, but like many in the NFC community who waited so long for Apple’s NFC endorsement, we know the world is ready for the next wave of smartphone IoT connectivity and look forward to playing a role in its future.
