A standard method for naming and addressing Machine to Machine (M2M) devices is going to be critical as these devices proliferate. A standard method will ensure that the devices are useful to the widest group of people, are interoperable among many networks and services, and propagate as useful tools for all.
There are many ways that M2M devices can access the Internet. Mobile/cellular, wi-fi, WiMAX, fixed, cable and satellite are the popular access networks. If there is a cost to access the network, the device will need to identify itself to the network before the network authorizes service.
This type of identifier typically is controlled and assigned by the access network and is called the access network subscription identifier. The International Mobile Subscriber Identity (IMSI) used in some mobile networks today is a good example of an access network subscription identifier.
M2M communication is typically between an M2M device and an M2M server. For example, a smart utility meter installed outside a house is configured to report the meter reading in an IP packet on a specific day of the month (e.g., end of the billing cycle) and time (e.g., low-traffic hours.) The information travels via the Internet to a server that then passes the relevant information to the billing system to generate the monthly bill for that house or customer. In the IP packet, there must be information about that specific meter so that the server/billing system can identify the associated customer account. This type of device identifier appears in the application layer and is controlled and assigned by the M2M service provider. There is a tight linkage between the server communicating with the device and the device. The server has an application that controls and manages the devices. This type of device identifier can be referred to as the M2M service provider subscription identifier.
When there is not a tight linkage between the server and the device, a different type of identifier is needed. The device must have an identifier that can be addressed by external entities and used by the access network to identify the intended device. This type of device identifier for the M2M devices is called the external identifier. Today the most popular external identifier for mobile networks is the telephone number, for the Internet, it is the domain name.
If you’ve been following this series you already know that telephone numbers are for people, not machines. There is industry consensus that for the long-term the M2M devices should not be assigned phone numbers and should have IP-friendly device identifiers such as the Uniform Resource Identifier (URI) , i.e., sip:user@domain; Network Access Identifier (NAI) i.e., user@domain; or Fully Qualified Domain Name (FQDN) i.e., abc.xyz.com. These types of IP-friendly device identifiers are globally unique and have large address spaces.
From the M2M service providers’ view, they would prefer that the same IP-friendly device identifier be used as the external identifier and the M2M service provider subscription identifier. Ideally, it would also be used as the access network subscription identifier. This would avoid having several identifiers for a device and mapping from one identifier to another. The external identifier and the access network subscription identifier should be controlled and assigned by the M2M service providers. This would support “M2M device portability” so that when the M2M service providers change the access network provider they need not change their device identifiers.
The industry has started to work on creating a standard naming and addressing scheme for M2M devices. This work is in its early stages and alternatives are still being evaluated. Once a standard is created, it will take a long time to implement in networks and devices. Given that most of these devices are using telephone numbers today, this solution cannot come quickly enough.
Comments are closed.