In the early days of the internet, researchers connected computers through simple packet switching networks to allow them to communicate. With research, this grew into new protocols that provide the basic capabilities that we have today—the web through HTTP, mail through SMTP/POP3, etc. But in the early 1980s, a graduate student at Carnegie Mellon connected a soda machine—using sensors for the status lights—to their computer gateway so that students—using the ‘finger’ protocol—could remotely know if drinks were available. This was the first known “thing” connected to the internet—or at that time, what was called the ARPANET, a pre-cursor of today’s internet—but it demonstrated the power of attaching more than just processing nodes to a network. The CMU device illustrated the concept and value of the Internet of Things (IoT) in a simple way and inspired what is today a market of over 18 billion internet-connected devices.
One of the key ideas in IoT is decentralization. Rather than centralizing compute and storage resources in a single place, they are distributed to where it makes sense. This decentralization was fueled by small low-power microprocessors that could handle the compute needs at the edges of the internet and NAND-based devices which provide fast high-capacity storage in a small space. Examples of these include the Raspberry Pi zero—an ARM-based IoT solution which includes Wi-Fi and Bluetooth Low Energy connectivity—and the Greenliant Flash Drive, which offers 8GB of mSATA storage.
Given the massive growth of IoT devices, new architectures for IoT solutions have begun to emerge. One particularly useful architecture is to proxy IoT devices using an intermediary device. In this architecture, an IoT gateway serves to mediate access to the internet for a larger number of IoT devices that may use other protocols—such as Wi-Fi, BLE, or other short-range communication schemes. The gateway also serves other uses such as acting as a secure gateway to local IoT devices and managing their provisioning, update, and overall maintenance. This hybrid architecture is particularly popular in industrial applications where the security of endpoint devices is crucial.
The prevalence of smartphones and broadband internet-connectivity also introduced changes to IoT device design. Home-based IoT devices can easily connect to Wi-Fi with enough bandwidth to support concurrent video and audio communication—using secure Wi-Fi connectivity solutions such as Texas Instruments CC3235x SimpleLink™. This allows devices to easily attach from smartphones to permit management and monitoring from anywhere in the world.
We’re now in an age of massive growth in IoT. According to Forbes magazine, by 2021, the IoT market will reach $520B, which is double the amount spent in 2017. But the future of IoT is not just about the devices that exist at the edge of the internet. The massive growth of devices will drive change into the overall ecosystem from cloud infrastructure to the means of communication among these devices. One of the largest growing segments in IoT will be in data centers and analytics.
One of the key areas that could provide friction to IoT growth is that of security. IoT devices continue to be a target of malware and black hats who seek fame or fortune in the vulnerabilities of IoT devices. The general problem with IoT security is that once a single device is exploited, a massive number of devices can then be exploited and used as a Botnet.
With ongoing research into security at massive scales and machine learning applied to these problems, IoT will continue to proliferate and find new applications in our homes, businesses, and cities. IoT will over time, become less about a market, and more about a fabric of sensors and compute capabilities that invisibly monitor and manage our world.
M. Tim Jones is a veteran embedded firmware architect with over 30 years of architecture and development experience. Tim is the author of several books and many articles across the spectrum of software and firmware development. His engineering background ranges from the development of kernels for geosynchronous spacecraft to embedded systems architecture and protocol development.