It was quiet in the Pod offices last week with what felt like half the team at MWC in Barcelona. As usual, it was an interesting, productive, hectic few days. The conference is enormous and there are always lots of exciting developments to discuss and people to meet. The biggest developments in the world of connectivity and the words that we heard most often as we crossed the conference floor were, ‘5G’ and ‘edge computing’.
Buzz has been building about 5G for a long time. This year has already seen a considerable amount of real-world testing. However, there are various technical hurdles still to be cleared and you probably won’t be streaming 4K video on your phone anytime soon.
What Does 5G Mean for IoT?
Most of the chatter about 5G has focused on the potential consumer uses. We’re more interested in what this technology means for IoT applications. One of the key IoT developments which 5G will facilitate is edge computing. The lower latency and reliable connections of 5G will enable edge computing to expand and develop. IoT devices will be able to exchange data between each other at the edge of the network without needing to transfer information to a central data system.
As this technology becomes more sophisticated, more data is gathered and normal patterns of use for individual applications become better understood, we will see a move towards machine learning and decisions being made at the edge. This will increase the speed and efficiency of these networks even further.
The Right Slice for Your IoT Devices
It is generally agreed that network slicing will be a key feature of the new 5G networks. With a 5G system, different bandwidths can be apportioned for different uses. So you can connect a smart meter to a slice with high latency and low data and reserve a different network slice with fast data speeds and low latency for self-driving cars.
Smart devices have different bandwidth needs. Network slicing gives wireless carriers the option of optimizing networks to serve the varying requirements of different types of IoT devices.
- High bandwidth/Low latency: self-driving cars, drones, robotic surgery
- High bandwidth/Medium latency: tablets, smartphones, VR/AR streaming
- Low bandwidth/Medium latency: IoT, m2m sensors, tracking modules.
Slicing the 5G network like this will reduce costs, increase efficiency and ensure the smooth implementation of a variety of IoT devices. 5G and network slicing will allow the IoT to move into its next phase of maturity.
Moving Away From the Cloud
After seeing a big shift towards cloud-based computing we will begin to see a move away from storing information on a centralized database. Intelligent devices on the edge of a network will collect and sort information at lightning speeds before making decisions on whether to act or if this information needs to be sent further up the chain. This will reduce the need for cloud storage centers
It will speed up a number of processes but could introduce an element of distrust. There could be a lack of accountability if complex (and potentially far-reaching) decisions are made at the edge of a network with no human oversight.
Once Again Security Will be a Serious Concern
It could have potentially huge security ramifications too. If devices at the edge are self-managing then how and when will a security breach be discovered?
This is why once again security should be an integral part of any new device or network developments. We need to design in security from day one in these new developments.
Unfortunately, many of these developments will not be strictly “new”. As we have seen in the world of industrial IoT new IoT technology is often added on to legacy devices, leading to a plethora of security issues.
As there are already so many IoT devices in the field the likelihood is that 5G connectivity, edge computing, machine learning processes and who knows what else will simply be bolted on to old and already insecure devices. Many of these devices do not support software updates or anti-malware and they could become the weak base on which whole IoT infrastructures are built unless security is taken more seriously.
However, by implementing security on the network you can have the ideal vantage point over your whole IoT system, allowing you to scale your application easily and economically. Pod Protect uses machine learning and heuristic monitoring to secure your network, making it the ideal choice for global security.
It was another interesting year at MWC in Barcelona which left us with plenty to think about. We’re looking forward to seeing how 5G develops and how IoT application developers utilize this new technology.