The power of network control in IoT connectivity.
From selecting a network provider to utilizing a Network as a Service.
Something that is often not considered when selecting IoT connectivity providers, especially in global projects, is network control and the benefits that can be offered from it.
With our mobile phones we are so used to the term international roaming, that we don’t think it could be any different with IoT. The truth is, though, that while each requires connectivity, mobile phones and IoT devices have very different needs. Mobile phones are powerful, complex devices that need to be charged often, connect to and browse the internet, and more.
IoT devices, however, are usually low power, aren’t intended to be charged often, and (ideally) have low computing needs — all of which makes for a poor matchup with most traditional cellular networks — which is where global providers come in.
Unlike traditional cellular networks, global cellular connectivity providers for the IoT industry can offer benefits such as increased security, improved device and data management and better overall customer experiences.
In this article we’ll run through the key considerations and benefits of network control to help you make the best choice for your IoT set up.
What is network control?
Network control is for example the idea that a device can stay connected to any network for any amount of time, no matter where in the world it is, without these other networks handling the data. Network control at a global level eliminates the need to rely on roaming partners, meaning that as a user, there are no time limits, extra charges, or security issues when using a network abroad.
The alternative to network control is relying on multiple providers to keep the same device connected at all times. For example, if your provider has coverage all across Europe — but nowhere else — you’ll have to rely on other providers to keep your device connected when you vacation in the U.S.
Why is network control important to IoT?
Network control isn’t something you think about with personal mobile devices because 99% of the time you’ll always have it in the same location, and when you don’t (for business, vacation, etc.) you sacrifice it for such a short period of time that it isn’t really significant and roaming typically suits your needs just fine.
IoT, however, is a completely different ball game.
Unlike personal devices, IoT devices are much more likely to leave their “home” network, and might even travel across the world indefinitely. And since most traditional networks aren’t designed for the global coverage needs of IoT, most projects end up having to give up network control for global coverage. This is also just as relevant for cases where devices remain in one place throughout their lifetime, but the producer wishes to distribute its devices internationally, rendering the idea of a “home” network equally unappealing.
But why does network control fix this?
When you’re trying to achieve global connectivity by patching the coverage of multiple providers together, it creates a massively complicated development process, full of differing contracts, rules, and regulations — all of which can change over time.
Not having network control also means that the burden of securing your data falls on your individual device, because data travels through providers you do not know and typically straight to the internet, adding complexity to the project and the devices themselves. This increases the risk of security threats, bugs, and transmission failures occurring. These kinds of errors can result in data being compromised or devices being lost completely, which is extremely detrimental to an IoT project’s longevity.
The current state of global connectivity.
Since the majority of service providers were created to provide connectivity to personal devices — like smartphones and tablets — they are far from ideal for IoT projects in need of global coverage.
In the current industry, there are three primary ways to achieve global connectivity.
With traditional cellular networks, there is next to only one way to achieve global connectivity, and that is through network partnerships. These partnerships are what allow devices to “roam”.
For example, say a device’s provider is Network A. Network A covers all of UK, so as long as the device stays in England, it will stay connected to Network A.
But if that device needs to go to Finland, then it’ll need to utilize roaming. This means that Network B — who has partnered with Network A — will provide the device with coverage in Finland for a certain amount of time.
While this works fine for most mobile devices, it’s less than ideal for IoT. For one thing, roaming usually has a time limit of 45 days or less. Meaning that if your device is roaming for 45 days without revisiting its home network, it will be cut off from Network B, forcing the device owner to partner with another provider while the device is across borders.
In IoT, it’s not uncommon for devices to cross borders and remain abroad for much longer than 45 days. And since global IoT projects could be made up of thousands of devices, switching them all to different networks over and over again is highly impractical, if not completely impossible.
A modern way to override these shortfalls is by using an eSIM platform that allows you to store and change multiple network profiles on each device. However, this system has its own complications that create different kinds of lock-in for your IoT connectivity.
LPWAN is a type of network connectivity that uses frequencies which haven’t been used in traditional connectivity. These frequencies previously haven’t been used because they transmit data at a slower rate; most of the things you use the internet for on your smartphone would happen very, very slowly over LPWAN.
This is enabled by the ability to send as small as possible data packets from the devices, which is a result of network control on the LPWAN provider’s side. The data is then encrypted once received by the network, which in turn makes the data ready to be transmitted over the internet.
There are two major competitors in the LPWAN industry: LoRa and Sigfox. Both use LPWAN technology in different ways to try and provide IoT projects with a low cost, low energy connectivity solution.
How data packets are minimalized using network control.
LoRa offers two way communication that transfers extremely small amounts of data with high cost efficiency. However, anyone trying to use this service outside of Europe is going to have to commit resources into deploying their own network gateway in order to get it off the ground. LoRa is also slower than Sigfox and is more susceptible to interference.
Sigfox works similarly to LoRa but is only uni-directional, and therefore has similar challenges. Because they are only just beginning to utilize LPWAN technology, the network is only available in certain regions. Unlike LoRa, though, they don’t have a DIY network gateway solution, so if you’re outside of a Sigfox region, you won’t be able to use it.
NB-IoT is a technology that uses the ideas of LPWAN — sending small amounts of data over a low-power network — and implements it on a cellular level. NB-IoT is faster than LoRa and Sigfox, and offers more reliable connectivity for connected devices. However, it doesn’t handle tower handoff very well, so it’s better suited to static devices, like meters or sensors. Furthermore, it is limited at a global scale until standardisation and roaming agreements are put in place.
What are the current drawbacks of LPWAN?
While all of these solutions provide IoT projects with an efficient, low-power connectivity solution, the global infrastructure isn’t quite there yet. They do offer different elements of network control, but are lacking in many of the other important areas that make global IoT viable such as scalability, availability and interoperability.
Global cellular + Network control.
This leaves us with the ideal solution, global cellular with total network control. This kind of connectivity combines the network control and low power transmission of LPWAN with the global infrastructure of traditional cellular.
These kinds of networks are known as global connectivity providers. They can offer connectivity in most major countries, eliminating the need for roaming and multiple service providers. These types of networks are relatively new, with most having been created specifically for IoT projects.
The impact of network control — Enabling Network as a Service.
Being able to have complete network control and the reliability of traditional connectivity has major potential within the IoT industry. It can improve your project’s security, efficiency, and reliability through the following services:
A great aspect of choosing a provider with network control is that they can offer you special features to help your IoT project run smoothly and more efficiently. One of the ways they do this is through the use of data connectors.
Data connectors receive raw data from the devices, adds the encryption and data endpoint specifications, such as required by Watson IoT, Microsoft Azure and others, before transmitting it over the internet securely. As a result, connectors remove the security processes and burdens from your IoT devices making them perfectly platform agnostic and flexible without any overhead.
This allows your devices to consume less power, reduces the complexity of the software required of your devices, cuts hardware and data costs, reduce the number of Over The Air (OTA) device updates required and more. This helps reduce the data overhead of your project, which is an important part of keeping IoT projects of all sizes manageable.
Additionally, updating data endpoints in the network is great when improving software features that require packets are sent change without having to update every single device OTA.
One of the major benefits of using a provider that has complete control over their network is that they can also provide customers with data authentication services. These services allow users to create custom criteria for the data that their devices send to the cloud, and vice versa.
Limiting what counts as “acceptable” data helps filter the data being sent between endpoints within your IoT project. This makes it harder for hackers to inject their own data into the project, which can have disastrous consequences.
In 2019, there was a hacking incident that highlighted the importance of data authentication. A hacker going by the name of L&M was able to crack into more than 27,000 smart cars. The hacker had the ability to track the location of each and every one of the affected vehicles and could have shut the engine of these vehicles off on command.
Fortunately, the attack was only conducted to point out the vulnerability of systems without proper security measures like data authentication, so no vehicles were permanently affected or shut off.
Another thing that network control offers is the ability to verify data integrity without any device overhead. Companies — IoT and otherwise — are under a lot of pressure from consumers and government regulations to keep data secure.
With IoT, though, this can be more difficult than it is for other industries. This is because every device presents a potential vulnerability in the network, an entry point for a hacker or even something as simple as corrupted/incorrect data. All the while the producer is incentivised to minimise cost and complexity of devices which limits their security oriented abilities.
Emerging tech like blockchain, though, can significantly reduce the risk of not discovering data that is potentially being tampered with. If it is, you can simply check the data from an earlier stage using blockchain. With complete network control, resources like these can be implemented throughout an IoT project. Blockchain does come with its own challenges, though, like storage and a lack of scalability, where the problem compounds when the device is required to handle it. These challenges are dramatically reduced using this kind of network functionality.
End-to-end encryption is one of the best ways to keep your data secure, even if it’s intercepted by an outside force.
Traditionally, end-to-end encryption is when your devices scramble the data they send using encryption parameters determined by your cloud platform. Then, when your cloud receives it, it is able to decode it using that same parameters. And when no outside force has that key, any intercepted data would be nonsense, keeping it secure and private.
There are two problems with this kind of end-to-end encryption. First, it means all of your devices have to have the ability to encrypt data. This increases the amount of time data takes to send, how much data is actually being sent, the processing power required to send data, and minimum required computation capabilities of your devices.
Second, there are different ways to implement security that need to be considered. How they are setup and reused within the network will require balancing need for stronger security of the network or lower device requirements. Selecting the wrong one could create a vulnerability that leads to bad publicity or reduce the feasibility of the IoT project.
When you are relying on multiple networks, this is the best you can hope for when it comes to end-to-end encryption.
When you have a provider that has network control, however, they can encrypt the data for you. This is possible because they are the only party handling your data between your device and the internet. This helps cut back on the cost, time, and device commitment of end-to-end encryption, all without a decrease in security and with no complex interdependencies between your device and your cloud.
The benefits of network control for IoT.
Data connectors, data authentication, data integrity, and end-to-end encryption are all services that provide your IoT project with the following benefits:
When you rely on a provider that doesn’t have complete control over their network, your data is going to be handled by your primary service provider and any networks that they’ve partnered with. This increases the number of ‘hands’ that your likely sensitive data is passing through, each being a chance for something to go wrong and increasing your device’s responsibility and complexity.
With a provider that has total control over their network — i.e., they aren’t relying on partnered networks to handle your data — your data is only ever passing between two parties: you and your provider.
This means that there is only one external party with your data, who can in turn add further security to your transmissions and data integrity, greatly reducing the risk of foul play and corrupted data.
Being dependent on a single global provider also means that you can more easily make changes within your IoT project. Rather than having to negotiate a set of features/needs with multiple providers, you only have to create one contract with your provider that can easily be adaptable to meet future potential needs.
Changes to your IoT cloud, databases, or underlying software are all much easier to account for with a single provider. Because data packets can be controlled at a network level, you can adapt to changes instantaneously.
This is important in any field, but especially in IoT, as the industry is in constant motion, as are the projects that fall under its umbrella. Having a single provider allows you to keep up with the fast pace of this IoT realm.
Having only one provider means that you can create a unique relationship with them that offloads the work required from your devices.
For example, as end-to-end encryption could be completed by your connectivity provider rather than your devices, this cuts down on the processor power that your devices require and the capabilities that your devices are required to have.
A single network provider also allows you to automate data overhead changes, reducing the need for OTA updates to your devices throughout their lifecycle.
Better customer experience.
And lastly, if your devices are being shipped to consumers, having a single provider will give them a simpler, easier to use product. OTA updates will be fewer, faster to implement, and less complicated. Your customers will also run into fewer bugs and negative software experiences. And when they do encounter these sorts of issues, resolving them will be a smoother experience with one party to interact with.
While global IoT connectivity is still in its infancy, as needs arise and problems are identified, network control will play a large role in how the IoT industry evolves. Thankfully, there are an emerging body of global connectivity providers to help your business stay ahead of the curve.
Onomondo is one these new global connectivity providers, and can provide your IoT project with a number of strategic advantages over traditional network options. We are currently developing our global Connectors so that you can have access to all of the aforementioned network services and more. Our aim is to support your business to improve security, efficiency, and flexibility, all while cutting overall costs and complexity. Get in touch to find out what we can do for you, both now and in the future.