IoT is a dynamic market that is not only driving change but responding to it. New applications appear daily, and the IoT ecosystem evolves along with it. Let’s explore the key factors that will drive IoT in the future and a couple of its challenges.
One of the largest segments of IoT is in the consumer space. Everything is becoming ‘smart’. From doorbells to locks and thermostats, IoT is bringing security and efficiency into our homes. What’s missing is integration and interoperability between these devices. While individual devices are useful, they will eventually inundate users with their own individual applications. The future of consumer IoT includes interoperability and integrated management that helps to simplify the use of the plethora of IoT devices that will enter our homes.
One of the main ways that IoT will benefit businesses in the future is through data. Data enables opportunities to understand how customers use products, which in turn can lead to new services or efficiencies in existing ones. One example is the use of video and machine learning to understand customer behavior or predict outcomes, such as interpreting expressions when viewing a product.
IoT sensors are an evolving area in IoT. Sensor technologies will move from simple, managed sensors that require processing by a local compute device to smart sensors that autonomously gather information and intelligently share information. This will include the ability for smart sensors to share information among themselves in order to reduce error as well as improve collection, detection, and prediction.
Telematics—or integrating location and communication with vehicle diagnostics and external sensors—is not new. But as IoT grows, particularly with traffic sensors and machine learning applied to vehicle diagnostics, so do the applications that can make driving more efficient and safer. High-speed communication technologies like 5G will introduce new capabilities, such as cloud-based entertainment or vehicle-to-vehicle communication for optimal traffic flow.
5th generation wireless will be a vital enabler of new IoT capabilities in the future. 5G provides greater bandwidth, lower latency, and higher density of devices within an area over prior technologies. 5G will enable IoT devices to not only communicate efficiently with cloud-based resources but also among themselves to support data sharing and cooperative processing (using spare computational or storage capacity). This will also improve data analytics, which will allow real-time optimization of IoT devices.
Machine learning is a key factor in the growth of IoT. The amount of structured and unstructured data generated by IoT devices could not be managed under human control. Therefore, machine-learning algorithms will gather and reduce the data to find its real value. This will be done at two levels:
As the massive numbers of IoT devices are realized, a few problems arise. Management and monitoring of devices at scale becomes non-trivial, and making use of the data provided creates bottlenecks. Machine learning, as discussed, can help. But beyond machine learning, technologies like sensor fusion can reduce uncertainty from collected data by fusing different sources. Autonomic computing can help devices become more self-managing and reduce cloud-level overhead in dealing with potentially billions of devices.
An area ripe for innovation and a point of friction for the future of IoT is security. This includes not just data security, but access security and overall management security of a potentially massive number of endpoint devices. A fundamental issue created by IoT is a massive collection of devices that share the same software. Once an exploit is found, it is then simple to exploit a large number with little additional work. This gives rise to Botnets and must be countered by autonomous management where the devices monitor and protect themselves as updates are created.
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.