Time to Ditch the Dumb Pipe - Part 1

How to compare IoT connectivity options

Choosing the right IoT connectivity option can be complicated!

Typically, it involves the weighing-up of five related factors which have to be evaluated, in combination and in relation to the needs of the relevant IoT use case.

1) Bandwidth

To state the obvious, different IoT use cases can have very different requirements, in terms of connectivity bandwidth. Security cameras streaming HD video will be more significantly more demanding than location trackers, sending a periodic update.  So, your chosen connectivity option must support the needs of your use case, at launch and over its expected lifetime. Available bandwidth can also be an important consideration, in relation to power consumption.  The longer an IoT device takes to receive and transmit data, the higher its power consumption will be.  Higher power consumption can lead to higher costs and/or shorter device lifetimes.  Many IoT devices don’t require high data throughput.  However, if you expect to deliver regular remote firmware updates to your devices over-the-air, data throughput may be an important consideration.

5G technology has been developed to support highly demanding, emerging use cases such as connected cars and smart-city and industrial IoT deployments, where both latency and capacity can be crucial. However, for the vast majority of today’s IoT use cases, non-5G connectivity technologies are perfectly adequate.

2) Coverage

How critical is your use case? For example, if a telecare device alarm is not sent, in real-time, the impact can be significant.  As a rule-of-thumb, the more critical the use case, the more important it is to deploy a multi-faceted approach, so that if the first connectivity option does not succeed, alternatives can be immediately and effectively employed.

For example, it is common for 2G/3G/4G mobile networks to be used as a fallback for Wi-Fi connectivity, to protect against power outages. In relation to mobile connectivity, it is not just the technology that it is important, the flexibility to access the radio networks of different providers from a single device can also be highly valuable. 

This can be achieved by having multiple mobile network identities (known in the industry as IMSIs and/or Profiles) on a single SIM card or eSIM.  If the IMSI of one operator cannot connect in a particular location or at a particular time, another can quickly be selected.  The same functionality can also play an important role in optimising the cost of mobile connectivity and for ensuring adherence to regulations in specific markets. 

We will explore these topics in more detail in Part 2 of this series, which focuses on Multi-IMSI and eSIM services.

3) Power consumption

This is an essential consideration for IoT devices that are battery-powered and not rechargeable. Improved power consumption can significantly extend device/solution lifetimes and thus reduce overall costs.

Wi-Fi wasn’t designed for IoT and is often therefore more demanding regarding devices’ battery power.  In contrast, LPWAN technologies like NB-IoT, LTE-M and LoRa have been specifically designed to save power and so extend IoT devices' battery life.

4) Security

This has always been an important consideration, particularly for more sensitive use cases. However, as we argued in our previous blog article How to Secure IoT (in the Network),  we believe that concerns over security and data privacy are now paramount in an increasing number of IoT deployments as the market matures.  A variety of complementary approaches can be deployed to ensure the integrity of an IoT solution, including zero-trust connectivity and security-as-a-service in the network layer.

5) Cost

Different connectivity options involve different charging models. There is no charge per MB or per GB with WiFi or Bluetooth.  However, mobile connectivity providers do charge for data passed over the upstream Mobile Operator’s networks.

For many IoT applications, which involve very low levels of data throughput, these costs are not material.  However, they can become significant as data consumption per device increases and/or there is a need to provide reliable coverage across a wide range of geographies or network providers.

Look for customer-friendly policies, like pooling data bundles across multiple SIMs/devices.  This will enable you to ensure that no data is wasted.  Look also at the levels of control that will be available to you regarding mobile network selection per SIM/device.  The more you can programmatically balance quality of service and cost in real-time, the more you will ensure that you are only investing precious margin where it is being well spent.

Serving future IoT connectivity needs

The needs of IoT customers are different to those of the traditional retail customers that mobile networks were originally built to serve. IoT deployments demand flexibility, scalability and real-time control and monitoring.  Enterprise customers have become used to working with cloud-based services that operate in exactly these ways.  However, the Mobile Operator technology stack has, by and large, remained inaccessible to customers and cumbersome to customise.  The increasing complexity of global IoT deployments drives a need for new technology platforms that empower customers with greater control of IoT connectivity services.

Increasingly, IoT customers will demand that the service they receive from their connectivity provider is well-tailored to their specific needs. Historically, such mass customisation has been impossible for Mobile Operators and their MVNO partners to support.  Customisation has involved a ‘project’ - projects that take time and involve high setup costs for IoT customers, resulting in many IoT innovations failing before they have even started.

IoT customers will increasingly want more control over the IoT service provided to them. For example, Enterprise IoT customers will require greater control over the security policies governing access to connected devices so that these devices can be managed just like another other assets in their IT estate.  In our view, IoT customers will want to combine these enhanced levels of control with more flexible, cloud-like commercial models - no hefty setup fees and a usage-based charging model with no minimums.

Traditional approaches to connectivity will no longer be able to meet the increasingly demanding requirements of the IoT market.  Optimising IoT connectivity at scale will necessitate new approaches, as IoT becomes increasingly critical to business and public services operations. As service providers, we must change the game by adding intelligent control as a new buyer choice dimension.

Intelligent IoT connectivity

In the early days of M2M and then IoT, cost and coverage concerns were paramount.  Now that the IoT market is maturing, the intelligent control of connectivity will begin to demonstrate its value in enabling more resilient, scalable and secure IoT solutions.

This article has explored some of the challenges and complexities involved in selecting the right blend connectivity options for IoT use cases.  Different individual technologies, such as 4G, NB-IoT, LTE-M, LoRa and 5G have clear pros and cons. However, this range of connectivity technologies is widely available from a range of providers, operating in a very competitive market.  There is relatively little differentiation between the portfolios of connectivity services provided by credible market players. Everyone is playing with the same set of tools, from a globally organised and standardised mobile connectivity ecosystem. 

Similarly, the economic model which has underpinned the development of global mobile network infrastructure has also served to maximise standardisation at the expense of innovation and flexibility.  The focus has been on building near-ubiquitous, high-bandwidth, reliable and secure networks to support the large-scale deployment of relatively undifferentiated services. In other words, dumb pipes have been sufficient, as long as they were wide enough!

Imagine instead what would be possible if every IoT solution developer or customer could apply the levels of control to mobile connectivity traditionally reserved for the Mobile Operators themselves, whether those developers and customers were using one SIM or a million SIMs. 

For some customers, it is fair to say, the benefits would not be particularly valuable – for some use cases, actually, the mobile connectivity ‘dumb-pipe’ is currently sufficient.  However, in our view, there is a growing proportion of IoT use cases where the customers’ lack of visibility over and control of the mobile connectivity service materially constrains overall solution performance and security.  In these cases, the ‘dumb-pipe’ mindset involves a significant and (as yet) hidden opportunity cost.

In the future, competitive advantage for customers (and forward-thinking connectivity providers) will not come from access to and the wholesale pricing of undifferentiated connectivity technologies but instead from the sophistication with which these technologies are controlled.  Intelligent IoT connectivity empowers customers with fine-grained, powerful control over IoT connectivity at the individual device level, in real-time.  This is the path to leveraging the true power of mobile connectivity technologies to enhance the value of IoT solutions.  It is indeed time to ditch the dumb-pipe!

Get in touch

Footnote:

It’s hard to provide a comprehensive list of all key IoT connectivity topics and features within a relatively short blog. So, if you think we have missed anything important or if you have different views, please reach out. We’d love to hear from you.