Today, most networked appliances on the Internet of Things (IoT) are not too demanding from a bandwidth perspective. Often they just share simple data that only require a few bytes to transfer from one device to another. For example, a sensor may update a computer with a machine’s current fuel efficiency, or the temperature of a commercial refrigeration unit – and, in network terms, this doesn’t add up to much traffic.
In fact, only around 1% of all network bandwidth is being used by the IoT. Yet, the 3G and 4G networks that largely support the IoT are designed for data-hungry connections, such as video-streaming services and web browsing. In other words, bandwidth isn’t the problem – it’s the price. The large U.S. wireless carriers require a certain Return on Investment to continue to build and maintain mobile networks and support the growth of IoT. Most IoT devices today provide a tiny fraction of the revenue that a typical smartphone generates in a month. The economics of IoT is a challenge for the wireless carriers today, and this problem will only grow over time as the number of connected devices grows exponentially in the next decade.
That’s why some people, like Google Developer Advocate Don Dodge, say we need a new, inexpensive IoT-dedicated network. Dodge argues that although Wi-Fi can be used to connect devices fairly cheaply, this is not practical because it limits use to the home or office environment, when in reality it should be available everywhere. Going outside Wi-Fi networks ranges forces users to rely on their data plans for broadband connection, which is not at all economical, especially considering the huge number of IoT devices we can expect to see in the future.
Dodge might well be right. Qualcomm, a U.S. leader in telecommunications and wireless products and services, is an example of a company that is backing a new long-range, low-power Wi-Fi standard (PDF). A network like this could be used to support the growing IoT by helping devices stay connected over a longer distance and lower cost, which could benefit business customers in the long run.
Others in the “new network” side of the debate come at the question from a different angle.
Existing networks can’t take the strain of a growing IoT
Cisco forecasts that the economic value of the IoT will reach a staggering $14.4 trillion by 2022 (PDF), and, as IoT gains momentum, we’ll see an increasing number of bandwidth-hogging devices connecting to networks. Wearables like Google Glass 2.0, the Apple Watch, Digital Signage — and a host of other applications and devices for mobile video — will continue to hit the market.
As more content is shared, we’ll see a need for constant and consistent Internet connectivity rather than the connect-as-required model we tend to rely on today. Some are therefore arguing that we will need a new network for the IoT to more effectively cope with the greater demands for increased bandwidth.
French startup SigFox sits firmly in this camp. The company has just announced that it is building a cellular nationwide network in Belgium exclusively designed for IoT devices. The company’s efforts could potentially solve spectrum and bandwidth problems by making sure that large numbers of IoT devices don’t overload existing networks.
But not everyone agrees.
Use existing networks to solve the connectivity problem
Some commentators argue the complete opposite: that rather than introducing new, dedicated networks, we need to fortify those currently in operation.
Suke Jawanda, CMO of Bluetooth SIG, argues that a distinct IoT network is unnecessary and is, in fact, counterproductive. Jawanda’s main criticism is that a distinct network would likely prevent the IoT from becoming a reality, because new network protocols could mean that machines would no longer be able to talk to each other. Jawanda strongly suggests that the three existing networks — cellular wide area networks, local Wi-Fi, and Bluetooth Smart Personal Networks — will be sufficient, but only if they are improved upon.
Dealing with the future of the IoT
There are a number of organizations and alliances where we will see continued debate on the matter. The IPSO Alliance, for example, promotes the development of “smart objects” and invests in companies developing those devices. The IPSO challenge is a competition designed to show “what is possible utilizing the Internet Protocol (IP) and the open standards of the IoT.” A cash prize of $10,000 isn’t to be sniffed at, but what this competition really does is drive innovation — and that’s what makes it interesting.
The IEEE — Institute of Electrical and Electronics Engineers — is another space to watch for news and developments. The organization is mainly focused on developing a comprehensive definition of the IoT itself and standards from which to operate.
Ultimately, though, a new IoT networking standard will either succeed or fail based on how many devices end up implementing it. With this in mind, the most interesting entity to follow in the world of IoT is Thread. Thread is a Google-backed consortium that’s currently developing a new wireless mesh protocol to supplement Wi-Fi and other current technology standards. Knowing Google’s affinity for the connected home, the seemingly limitless resources it has, and the number of companies already represented in Thread’s membership ranks, my money’s on these guys.
Rob Chamberlin is cofounder of DataXoom.