There are several technologies available to connect objects, some of which are household names, some of which are less well-known. We currently go from Bluetooth to ZigBee to Wi-Fi for personal and home environments and from Sigfox to LoRa for longer distances. But connectivity through SIM cards also has a role to play, in particular for mobile connected devices. Mobile connectivity is by definition pan-European. Connected objects do not stop at borders. The political agreement reached in June on the abolition of roaming charges by 2017 is a great step forward but more needs to be done in frequency harmonization, numbering and addressing.
Take one example, the car industry. Innovation in the European car industry is already opening new horizons with applications for so-called "connected cars". In the same way as a chassis number identifies a car regardless of its physical location, and no matter where the car is produced or sold in the EU, we should have unique digital identifiers for connected cars regardless of their location in the EU. I therefore think that is time to discuss how a combination of European numbering and IP addressing can deliver a practical and effective solution.
New business models, new solutions
Many operators have realised that and are working on their service strategies. But we still have some unanswered questions in the Digital Single Market:
- Can our current numbering systems cope with the potential of billions of connected objects being produced in one Member State but used throughout the Digital Single Market?
- Will Europe be able to offer business opportunities for Machine to Machine (M2M) services that are truly pan-European or global by nature, while the unique identifiers required to locate the relevant objects correctly are managed by individual Member States?
It seems to me that such new business models require us to consider innovative numbering solutions to accommodate the requirements of M2M customers and service providers, as well as the volumes of numbers at stake. One key question, for example, is how to avoid M2M service providers becoming locked in to a single provider, simply because of the prohibitive cost of switching?
I am also aware that the NRA-administered number schemes for mobile phones (so called E.164 and E.212 numbers) are by no means the only ones that are being used to uniquely identify objects in networks. In fact, other industries such as automotive, avionics and retail already track objects in their global supply chains and their specialised numbering schemes are implemented in IT systems around the world. So there is more than just one starting point to tackle this challenge.
In addition, we need to consider that all these billions of objects will have to be connected via the internet in order to gather data or remotely to enable value-added services. It is quite fortunate that, thanks to the Internet Protocol version 6 (IPv6), the Internet Engineering Task Force (IETF) has provided a global solution for the long-anticipated problem of IPv4 address exhaustion. IPv6 provides a much larger addressing space to ensure that every device on the Internet is assigned an IP address for identification and location definition. There are still limitations with mobility management, but I expect to see positive developments in the future in terms of IPv6-based true mobile connectivity. In the meantime a pragmatic and effective solution may be to combine numbering and IP addressing.
In my view, this is a very important challenge for the development of M2M services in the Digital Single Market:
- How can we ensure that billions of objects which either do not have a predefined static location (e.g. because they are tracked all the way from production through usage and on to recycling) or which are permanently mobile (e.g. connected cars, fishing vessels, etc.) can be connected via the internet?
Considering the significant investments made by various industry sectors in their numbering schemes, as well as the capacity of the IPv6 addressing space, it could be that there is no one solution that will fit all cases. Instead, we might have to consider a DSM numbering scheme that brings together specialised unique identifier systems with IPv6 addresses to facilitate a seamless integration into the internet. I see an innovative European numbering scheme as an important element in establishing a European "Internet of Things" ecosystem.
The currently fragmented regulatory landscape responds poorly to the needs of increasing deployment of M2M services. The lack of regulatory certainty related, for instance, to interconnection, the risk of an exhaustion of national mobile numbers, the potential confusion created by permanent use abroad of national numbers, and the diversity of national numbering regulatory requirements create unnecessary obstacles.
A European numbering scheme could provide additional numbering resources with harmonised requirements and enhanced enforcement possibilities but without negatively affecting Member States' ability to manage their numbering resource. On the contrary, national regulators would be responsible for the assignment and for collectively managing such a numbering scheme.
A European Telephony Numbering Space (ETNS) was less needed few years ago. It was not a great success, I must say. Perhaps it was a solution looking for a problem. But now the issue of a single European numbering plan has become relevant again with the rapid development of M2M communications and large scale pan-European applications like eCall. One car with its own chassis number, unique number and unique address.
In the forthcoming consultation on the review of the Telecom rules I look forward to hearing your views on numbering schemes and the interplay between connected objects and how you see the emerging needs for a European level solution.