Internet of things was first discussed in 1990, but the impact of this emerging field is only being felt now. The internet of things (IoT) is still shaping the way we manufacture products, manage buildings, and collect data from our surroundings. By bringing intelligence to the endpoint devices, we can process and understand the big data we collect at a whole new level. This translates into better functioning systems, such as better healthcare systems that have the power to focus on preventive care, more efficiently managed buildings, and better quality of living. Still, the rapid adoption of IoT devices has surprised us all with its magnitude. IoT analytics estimates there to be more than 27 billion IoT connections by 2025. The rapid adoption of IoT technology has caused an interesting problem: Can the existing internet protocol is known as IPv4 accommodate enough IP addresses for all internet users once we have 10 billion additional devices connecting to the internet?
Limitations of the Existing Internet Protocol
The internet protocol (IP) aims to work as a network layer communications protocol for transferring datagrams across various network boundaries. The Internet protocol hereby is the fundamental building block that establishes the internet by enabling internetworking. The current network protocol, IPv4, was developed by the United States Department of Defense and first launched for production on SATNET in 1982 and fully deployed in Advanced Research Projects Agency Network the year after. IPv4 relies on a 32-bit address space, which provides a total of 4,294,967,296 (2³²) unique IP addresses.
When available IP addresses are limited to roughly 4 billion unique identifiers, we can clearly see that deploying an additional 10 billion IoT devices can cause issues. Businesses have already found workaround solutions to this problem by implementing network address translation and other technology layers that enable multiple locally connected devices to share a single address. These types of connections are called subnets.
IPv6 and the IoT Revolution
Organizations and cities worldwide are increasingly interested in harnessing the power of IoT devices and endpoint AI. IPv6 has collected plenty of interest from the business community, and in November 2021, Google measured a 33.55% IPv6 adoption rate globally, with countries like India, Malaysia, and Taiwan leading the curve. IPv6 protocol has been proposed as the potential solution to the IP address shortage. IPv6 addresses are 128-bit long, which could, in theory, accommodate nearly 340 undecillion unique IP addresses, thereby leaving enough room for billions of devices and internet users to connect.
How Many Devices We Will Need to Accommodate?
According to Statista, worldwide end-user spending for IoT solutions will reach $1,567 billion by 2025. As businesses are looking for more ways to gain visibility to their processes through data, and consumers are increasingly invested in taking matters like health into their own hands, demand for wireless endpoint devices is increasing. In the near future, we will have intelligence embedded in every last piece of lab equipment and factory machinery in the world. For the internet of things and endpoint AI, the time is now. We will see billions of devices connecting to the internet and making our lives more productive, data-driven, and convenient.
How Ambiq Helps
Ambiq® will be enabling endpoint AI for billions of wireless IoT devices worldwide. Calculations for AI inferencing are power-hungry, but with Ambiq’s ultra-low power solutions, based on proprietary Subthreshold Power Optimized Technology (SPOT®), we can reduce endpoint AI power consumption on the order of nanoamps.
Simply put, our solutions enable endpoint intelligence everywhere.