Three’s Not a Crowd: 5G, IoT, and Edge Computing

IoT devices are on a rapid growth trajectory, revolutionizing homes, cities, and places of work. However, their potential is being held back by a looming network data bottleneck — and 5G is our most promising solution.

Right now, smart devices are getting smarter: A refrigerator alerts its owner when milk is running low, and smart TVs serve as central hubs for smart homes equipped with IoT devices. Demand is also rising for commercial applications of IoT; smart factories fitted with sensors assess productivity and output, and retail store sensors monitor customers, employees, and goods.

As IoT devices and their applications grow more complex, they send increasing amounts of data to the cloud. However, there is a limit to how vast the IoT ecosystem can grow without introducing a data bottleneck in the network and then, latency. To accommodate a massive uptick in IoT data, the industry first leaned on edge computing, pushing data processing and AI capabilities out from central cloud servers into further-flung parts of the network — or even IoT devices themselves. But this approach is reaching its limit. As complexity grows, a larger network is required to support edge computing and IoT devices, and it's time for 5G to take the reins.

5G and Edge Computing

In edge computing, AI processes data at the device or edge server level, rather than in the cloud, significantly accelerating response time. Consider popular home surveillance cameras on the market record high-definition video 24/7. It would require a massive amount of bandwidth to ingest and process this volume of video data in the cloud, introducing high latency. Instead, home surveillance technology relies on edge computing to record and send only relevant footage (such as movement) to the cloud for user review.

Edge computing tied with IoT devices made the idea of smart homes and smart factories possible, and the related data more manageable to process on a mass level. However, as IoT device ecosystems become more complex and grow in popularity across both private and commercial applications, edge computing needs a bigger network. The technology relies on network access to receive the important algorithms, machine learning, and processing models necessary to operate. Without additional network bandwidth, it would be incredibly difficult to support a society that relies on increasingly complex IoT devices. Just as IoT needed edge computing to achieve the next level of connectivity, edge computing needs 5G to support increasingly complex applications.

How 5G brings IoT together — and introduces a new set of roadblocks

5G networks widen the pipe that carries data to the cloud, allowing for significant increases in data transfer and speed. This results in the low latency of 5G networks that makes sub-millisecond response times possible.

For applications that rely on the cloud, like gaming, 5G’s rapid response time is critical for user experience. But 5G can also advance more sophisticated IoT applications, like self-driving car technology that relies on zero latency to safely avoid obstacles and trigger the brakes. This level of latency was simply unattainable when applications relied on 4G and LTE.

But is 5G the definitive solution to increased operational efficiency and low latency? In some cases, no. Many on-premise technologies lack the processing capacity to efficiently handle the large streams of incoming data from 5G’s widened pipe. In these cases, 5G reintroduces latency. That’s an annoyance and a productivity issue for industrial applications, but it renders the IoT device unusable in others like healthcare, where any latency in applications such as remote temperature monitoring could put patients’ health at risk. Looking ahead, such applications will require large-scale infrastructure and technology upgrades to harness the full potential of 5G.

Full-scale 5G for IoT requires infrastructure that can handle the data

Full-scale 5G network adoption for IoT devices will require significant infrastructure investment and upgrades, which in some cases may not be feasible or logical. The industry must also consider the need for conversations and education around privacy and security implications that come with an increase in sensors and cameras in both homes and workplaces.

As with edge computing, 5G is essential for next-level IoT connectivity. Despite setbacks and necessary investments, building a world with full digital harmony among devices, factories, homes, and cities is a possible reality — and we are well on our way to that future.