Join IoT Central | Join our LinkedIn Group | Post on IoT Central


New Developments in IoT Connectivity

New Developments in IoT Connectivity

 

Guest post by Peter A. Liss.

Connectivity is wrongly thought of as a commodity, including in the IoT context. This article will give an overview of current developments in IoT Connectivity, and look at their effect on Network Operators, Platform vendors, IoT Solution Providers, and Enterprise & Consumer customers. 

I also cover the likely impact of 5G, Narrowband IoT and programmable SIM cards, and SDN (Software Defined Networks). These new connectivity technologies will bring differentiation, innovation and new revenue from IoT.

OVERVIEW – CONNECTIVITY AND DIFFERENTIATION IN IOT

These new IoT developments include:

1.   Newer networks such Sigfox, LoRA, Narrowband IoT, and soon 5G.

2.   IoT platforms that can manage all types of connectivity.

3.   The growth of eUICC (e-SIMs) or programmable SIMs.

4.   IoT connectivity platforms using SDN (Software Defined Networks).

There are two opposing views about connectivity. On the one extreme, some Vendors pitch that “IoT Connectivity is the foundation of differentiation” (recent Ericsson Webinar). At the other extreme, some Enterprise customers buying these services assume “all IoT connectivity is the same”. 

In my view, the truth is in the middle. On the one hand, IoT hardware such as sensors and IoT applications could drive even bigger differentiation and innovation than the type of IoT connectivity. On the other hand, IoT connectivity should never be viewed as just a commodity that is plug and play.

HOW TO DIFFERENTIATE WITH IOT CONNECTIVITY:

Let’s take a closer look:

1)   There are many different types of Connectivity to choose from (cellular, WiFi, Zigbee, Satellite, and different types of LPWAN (Low Power Wide Area Networks). The criteria for selection include data cost, device cost, data rate/speed, battery life, outdoor and in-building coverage, and latency. Some of the much talked about networks like 5G are not yet available, and Narrowband IoT is in its infancy.

2)   The variety of connectivity offerings are increasing. Even taking a single technology like 4G, the offerings in terms of coverage, cost, roaming, integration effort, and customer service do differ widely.

3)   Costs are declining– the cost per MB has decreased, however, this is not the same as connectivity being a commodity (i.e. indistinct service). On the contrary, with more offerings and price competition, there is a greater need to choose the connectivity provider carefully. Pricing models may differentiate not only on cost per MB, but also with additional charges for VAS, the period charged for (monthly, per annum etc.) or number of connections included, or amount of data included in a packaged price. In the case of LPWA, charging can be per message, and not just per MB.

4)   The IoT Connectivity platform is where some of the disruption is happening. This platform manages the cost of connection, quality of service, SIM and device status. Along with the type of connectivity chosen, hardware (gateways & sensors), and IoT Applications built, the connectivity platform will be a key differentiator to your business case or service launch. 

My scheme below shows the place of the IoT Connectivity Management platform as the foundation of the IoT technology stack. Some differentiation could be achieved at any level in the Stack, but the effort required to offer a total solution will depend greatly on the Connectivity chosen at the bottom of the stack.

0?e=2119644000&v=alpha&t=zIRICyRP4vgTyqeq_nHh69LnmGHT-ahawOYS3Mp_uDQ

WHAT USER CASES WILL NARROWBAND IOT SUPPORT?

Narrowband IoT (NB-IoT) greatly improves network efficiency and spectrum efficiency and can thus support a massive number of new connections. The same is true of the sister technology Cat-M1 in US, which may also play a role in Europe in future. The majority of these new IoT connections will be industrial IoT (IIoT) solutions that require long battery life, and ubiquitous coverage (including remote areas or indoors). These user cases also require competitive pricing models for low bandwidth solutions, since many industrial IoT cases are not data hungry. 

Some examples of Industrial use cases are monitoring of oil and gas pipelines for flow rates and leaks, noting that often there is no external power in inaccessible areas. Warehouses are another industrial user case for tracking goods with pallets equipped with an NB-IoT module. NB-IoT modules have a long service life, require no maintenance and have a link budget gain of 20 decibel compared with a conventional LTE deployment, giving approximately 10x more coverage than a normal base station, thus penetrating deep underground, and into enclosed spaces indoors. 

Consumer examples of NB-IoT are luggage tracking (click for link to Sierra Wireless Case study), air quality monitoring, and children’s communication devices, and parking solutions.

NB-IoT, is a software upgrade to existing cellular Base Stations (or if the Base Station is old, a new circuit board must be inserted). The Core network also needs some upgrading. NB-IoT is reliant on a SIM card in the IoT device/gateway and partly because of the SIM it offers the same security & privacy features expected of cellular networks. LPWA technologies, such as NB-IoT and category M1 (LTE-M), also offer increased network coverage over a wide area, at a low cost, and with very limited energy consumption. In the case of Narrowband IoT, a battery life of over 10 years or more, is promised by Vendors (it remains to be seen - in the field, it might need a larger battery at an extra cost of approximately 20 Euro).

NB-IoT networks are already becoming available, for example, Deutsche Telekom has rolled out its NB-IoT network to approximately 600 towns and cities across Germany since launch in June 2017. According to Telekom, more than 200 companies now trialling the technology already via commercially available test packages. Nationwide rollout in the Netherlands was completed in May 2017 and Deutsche Telekom brought the technology to six further European markets by the end of 2017. Other major operators have similar roll outs for NB-IoT.

As expected, many IoT platforms are now being designed or upgraded to offer Narrowband IoT connectivity management. Cisco already announced in 2018 the availability of NB-IoT on its Jasper Control Center platform.

WHAT WILL 5G BRING TO IOT?

5G is not yet available commercially, and we can expect the first roll-outs in selected countries in 2019, and even then, just city coverage, or home-based 5G. High speed, high reliability and low latency are the main benefits of 5G.  Whilst NB-IoT is targeted specifically at the IoT Market, 5G is targeted at business & consumer users too. Also, worth noting is that the NB-IoT roll-out is ahead of 5G.

Regarding the high bandwidth of 5G, example uses include security cameras and monitoring, computer vision used in Industrial production, connected car user cases (infotainment, autonomous vehicles, and safety), and traffic control in Smart Cities. The increase in speed between 4G and 5G can be as much as 100 times. This makes a big difference to user cases that require uploading and downloading of video-based content faster and in larger volume.  It remains to be seen whether IoT applications will need to use such high data speeds. Perhaps it will be the Augmented or Virtual Reality cases (AR and VR) that utilise this bandwidth.

With 5G there is very high reliability, which is important to support mission critical services in IoT (e.g. medicine, industry, traffic control). However, the real benefit for IoT is likely to be with the low latency of 5G. Low latency allows more of the computer processing or data analysis required by a device (IoT Gateway or Smartphone) to happen in the cloud. With latency of under a millisecond, there is almost no difference that the data is processed in the cloud rather than the device. This has perhaps more implications for the IOT Solution architect, rather than the user.

Indeed, the user cases that depend on 5G’s low latency are still to be proven in practice. For non-IoT user cases (i.e. human interaction), the latency (such as changing of a pixel on a TV, or response time for instant messaging and online Presence) might not be noticed. However, for an M2M or IoT application in theory there is a great need for low latency and a machine might notice the difference in latency when a human does not. For this reason, the low latency is being pushed by the 5G industry as compelling for IoT (but yet to be proved). IoT user cases that are expected to benefit are remote industrial control, and autonomous vehicles, where milliseconds could be critical.

As explained in the discussion of latency, one change with 5G could be more processing in the Cloud, especially with Edge computing being a focal point in the architecture, and this might help reduce 5G IoT device prices. Other Emerging developments that might affect IOT include virtualised RAN (Radio Access Network) and network slicing. Virtualised RAN is intended to offer bandwidth with lower network costs, since by “slicing” the RAN, it is not necessary to utilise the whole core network, but rather allocate a part of it and the associated costs, thus allowing for profitable use cases with 5G.

WHAT ADVANTAGES DOES A PROGRAMMABLE SIM OFFER IN IOT?

Programmable SIM cards (also called eSIMS or eUICC ) are not new. What has changed is the number of service providers that offer them for IoT. Some prominent examples are Stream, EMnify, Cubic Telecom, KORE, Nokia WING and Teleena. Furthermore, the new generation of Smart SIM and associated management platforms are challenging the MNOs in terms of quality of service and signal coverage. They might also challenge MNOs in terms of cost - see the section below on SDN.  

The “e” in eSIM can mean both electronic (it can switch network and be programmed over the air) and embedded (i.e. deep inside machinery, a car or a remote location). In other words, you do not need physical access to the embedded SIM to update it or to change network, service or security settings.

The advantages of an eSIM are that it can be programmed over the air to find the strongest signal, or according to customer network & service preferences. When a data-service failure is detected, the eSIM can switch dynamically to the best network service. Consider a user case such as Smart Metering. The meter is always connected by being programmed not only to select the strongest signal, but also to select the signal that is best for your Meter technology and customer requirements.

In sum, the IoT Service Provider does not own a network, but can still offer the following to its customers:

•Issue own SIM cards, that can be embedded and switch operator over the air.

•Attach to the best or cheapest radio signal (RAN) – automatically

•Billing capabilities, often in real time, for the pricing of new IoT services.

WHAT IS THE IMPACT OF SDN ON IOT?

As explained above, the e-SIM is the first disruptive step to being able to offer an IoT service, without being tied to one specific radio network (RAN). The second step is to bypass the Operator’s core network. This is now possible with some Service Providers using Software Defined Networks (SDN) and NFV (Network Feature Virtualisation). They have built their own virtualised core network that is cloud hosted. EMnify is one example that can offer the following advantages:

•Low cost, because designed for IoT, and using proprietary technology (therefore no licencing costs)

•Auto-configuration and scaling. Because it is Cloud Based the service is truly elastic (i.e. can be quickly and simply expanded to meet customer demand for increased data volume, or larger number of SIM cards)

•Pay-as-you-grow pricing

•Flexible and Real time billing that is accessible online

•Have own numbering resources (IMSI, IPv6, MSISDN)

•Manage your own virtual mobile IoT network including Elastic Packet Core, Subscriber Management, OSS/BSS, Management Portals and open APIs. 

•A private and secure device cloud and implement own security policies (such as own VPN – virtual private network - in the core network in the cloud).

The “Gorilla” MNO (e.g. Telekom, Verizon, Vodafone etc) is reduced to providing only the radio network, and with the eSIM you can actually switch networks. To be clear, you are not reliant on the operator for the core network at all, and you have a choice of radio network. In sum, the advantage is that such a virtual network in the Cloud allows IoT user cases that have lower revenues, because the IoT platform is designed for lower connectivity costs.

 

CONCLUSION – DISRUPTION IN THE IOT CONNECTIVITY MARKET

I have built the case that “boring” connectivity is going to be disruptive for IoT, and it will generate growth. In sum, this is because many IoT business models require lower costs for the lower “micro” or “mini” ARPU/revenue that they generate. Secondly, these new network technologies bring improved speed, latency, battery life, and coverage. Thirdly, new technologies like eSIM and SDN, give the customer choice and independence from the MNO.

Enterprise customers will need to get more knowledgeable about the types of connectivity on offer, and the pros and cons, and costs of each technology. Disruption in the market is starting, due to many new offerings from MNO, and MVNOs that are IOT focussed. 

Price declines for NB-IoT and 5G enabled devices will also be business drivers. Many connectivity platforms will struggle to distinguish themselves, but can do so, for example by focussing on particular Verticals, or a specific geographical focus, or own Cloud-based packet core. Enterprise customers need to get the balance between a price that enables the business case, but also choosing connectivity that provides the best service level. 

LPWA technologies such as Narrow-Band promise to open-up new business models due to lower device and connectivity costs better coverage and longer battery life. NB-IoT is still in its infancy and these benefits like lower device costs are still to be proven.  Importantly, the connectivity costs of NB-IoT (as well as module/device costs) will need to be low enough to support the proposed new business cases like parking meters, water meters, luggage tracking, pipe monitoring, and tracking goods in warehouses. 

5G for IoT will enable data hungry business models, insure against capacity constraints, and provide wider coverage and almost no latency. Since 5G roll-out is still in the future, it remains to be seen if (or when) the required network density (using such small cells) is enough to provide the wider coverage and higher data rates promised. Almost zero latency is likely to be the most interesting feature of 5G for the IoT World, especially for critical applications like autonomous driving and industrial control.

Big data, Analytics and Application Enablement Platforms/AEP might sound more exciting and promising for innovation and differentiation in IoT. They sound more compelling than a connectivity management platform and new types of connectivity. However, Connectivity is still the foundation of the IoT business case. It is not a commodity. In particular, Narrow-Band IoT, eSIM and SDN will drive new growth in IoT, together with the imminent roll-out of 5G.

Copyright: Peter A. Liss, an independent and commercially focussed IoT expert, based in Germany, who is also available for freelance consulting work.

This post originally appeared here.

Cover photo by Federico Beccari on Unsplash

David Oro is the Editor at IoT Central

Sponsor