As global organisations increasingly adopt IoT solutions, Peter Van Den Houten at Kore discusses the major innovations set to transform connectivity and networking infrastructure
It is perhaps an obvious statement to make, but the truth of the matter is that a lot of the fixed and wireless network infrastructure that serves IoT deployments today is somewhat legacy but certainly not purpose-built nor best suited to the demands that advanced IoT applications bring.
As we’ve transitioned from traditional machine-to-machine applications to those that can be considered true IoT, the demand on the data has grown exponentially.
We are no longer dealing with devices that simply ‘check in’ once a week and if they don’t, it isn’t the end of the world. Now we’re dealing with devices that depend on reliable, fast connectivity to transmit their data in near real-time.
However, that is merely one part of the story and the other half, which is often forgotten, is the back-end network infrastructure serving the critical systems that rely so heavily on the data coming in from the devices.
Every IoT device you deploy is an extension of your network and if just one of those devices becomes compromised the outcome could be disastrous
The story is similar, as corporate networks have been designed with humans in mind and connectivity has been optimised to ensure that end users can access the applications and systems required on their laptop or smartphone.
Simply put, IoT applications function in a fundamentally different way. The level of data consumption is relatively low compared to regular IT applications, but the way in which the data needs to be served up brings with it new demands for connectivity that provides extremely low latency, quality-of-service and built-in security.
Another important consideration is network security, which is often an afterthought with IoT but is perhaps one of – if not the – most important factor to consider. Security for ‘traditional’ network infrastructure and its end users is a known beast nowadays but in many ways, it is still tossed to one side when it comes to IoT applications.
The reality is that, in some way, every IoT device you deploy is an extension of your network and if just one of those devices becomes compromised the outcome could be disastrous.
Finding the right network solution
One of the most important considerations at this point in time is the introduction of new, licensed low-power technologies, such as NB-IoT and LTE Cat M1, which are purpose-built for IoT applications.
Related to this is the phasing out of ‘legacy’ 2G and 3G technologies. If you’re operating on a local scale, this challenge is relatively easy to address, however when we look at this from a global perspective, the question becomes a great deal more difficult to address.
For example, in the United States and Australia 2G has already been phased out and roll-outs of NB-IoT and LTE Cat M1 are well underway. Compare that with Europe where 2G is likely to be around for a while yet – in fact it is probable that 3G will be phased out before 2G – and whilst many NB-IoT deployments have been carried out, coverage cannot be considered comprehensive.
Simply put, trying to find a ‘one size fits all’ network solution to serve your customers for the next five to ten years is a tall ask and the decisions you make and the network technologies you choose to support are all-important.
Before you make any decisions, make sure you have a good understanding of the cellular technology landscape in the regions that you wish to deploy to, as well as the ‘migration’ paths available to you as and when 2G and 3G are phased out.
Looking towards 5G
The exciting thing about 5G is that, whilst it isn’t specifically purpose-built for IoT applications, the multiple enhancements it brings over 4G will potentially ignite a vast number of IoT use cases that have, so far, remained constrained by technology.
One such enhancement is the rate at which data can be transferred and the extremely low latency that comes with that. Currently, IoT applications that rely on 4G transmit a relatively large amount of data, which takes time to process. In the case of the autonomous car, where there is a heavy reliance on real-time data to make split-second decisions, it just doesn’t fit the bill.
The rollout of 5G will bring with it networks that are capable of data transmission times that are up to ten times faster than today’s 4G networks, meaning data can be transmitted and therefore processed much quicker and in turn decisions based on that data can be made in real-time.
New infrastructure means there is heavy investment required by the mobile network operators
5G will not only support the advancement of cloud-based data analytics for IoT but it will also support lightening fast device-to-device communication. Continuing the example of the autonomous car, this will make all the difference in preventing collisions and improving driver safety.
Preparing for change
The jump from 4G to 5G is more than just numeric – there are costly infrastructure upgrades required to make 5G a reality.
The key difference in comparison to previous upgrades, from 2G to 3G and then from 3G to 4G, is that 5G needs to operate at a frequency that provides much higher bandwidth to what is used today for 4G, meaning that brand new infrastructure is required.
Operating at a higher frequency also means signal propagation is lower, so more of that infrastructure is required in any given region to ensure coverage, although that infrastructure will consist largely of small-cell base stations that are strategically placed to optimise coverage.
New infrastructure means there is heavy investment required by the mobile network operators and it will naturally take some time before 5G networks can be considered widespread.
The reality is that new cellular technology rollouts take time. Typically, the shift from one technology to another works on a ten-year cycle and the added complexity with the cellular ecosystem is that there are so many parties that have to come together to agree on a common, standards-based approach to solving the many technical aspects that a new network technology brings.
So, with that said, the ‘good’ news is there is plenty of time to plan.
The key consideration with 5G is that it is going to deliver radical change to the way that IoT applications can operate and the use cases they can meet. This isn’t just a case of ensuring that IoT applications are connectivity future-proof, it’s a matter of defining a strategic roadmap that enables the IoT applications to evolve and tap into the 5G feature-set as and when it arrives.
This approach needs to extend beyond just the hardware and the connectivity layer, it needs to include transformation of the data processing layer too because 5G truly has the ability to allow the device, network and application layers to function in unison.
To read more around cellular trends, download the latest Kore whitepaper ‘Guide to European Celullar Technology & Trends’ here.