  | Issue: | Global 2013 |
| Article no.: | 11 | |
| Topic: | Challenges in handling M2M devices in permanent roaming | |
| Author: | Tal Meirzon & Shlomo Wolfman | |
| Title: | CEO / Co-Founder, CTO | |
| Organisation: | Starhome MACH | |
| PDF size: | 195KB |
About author
Tal Meirzon was appointed CEO of Starhome MACH in December 2012. Prior to joining Starhome MACH, Mr. Meirzon served as the CEO of Wavion Wireless Networks. Tal Meirzon holds a B.Sc. in Electronics Engineering and an MBA.
Shlomo Wolfman is the co-founder, CTO of Starhome MACH. Previously Mr. Wolfman led development teams at companies such as Comverse Technology, VDOnet and Tadiran Telecom. Shlomo Wolfman holds an MBA and B.Sc in Electronic Engineering.
Article abstract
The technology and the value proposition of M2M has long been a focus for many mobile operators. Now they are faced with a new range of challenges in the area of mobile connectivity for M2M devices; from managing permanent roaming devices in the network, to the visibility of M2M devices and signaling optimization.
Full Article
A large number of M2M devices are now roaming on a permanent basis. These include vehicles that are manufactured with built-in SIM cards and shipped to various countries around the globe. In fact, most M2M equipment, such as cameras, smart meters and health care equipment is manufactured in one country and distributed globally. One of the main issues here is that an embedded SIM card cannot be manually replaced with a local SIM, meaning the M2M device is connected to the visited mobile network as a roaming device, which translates into higher costs for operators due to increased outlay to cover signaling.
To effectively tackle the issue of permanent roaming devices, industry players are currently working on the dynamic ‘IMSI swap’, the technical solution for remote download of a new mobile identity. This solution will enable an M2M device to move from its home network and become a local device in its new roaming network. However, this technology has not yet been deployed. In an effort to solve this problem there are two industry approaches – those who are pushing the programmable SIM solution (eUICC), and those who are pushing towards a TRE-based solution (Trusted Environment) in the device itself.
Visibility of M2M
One of the challenges faced by mobile operators is to identify which M2M devices are receiving services from their network so that they can apply the relevant tariff. Based on identification, the operator is able to analyze consumption of inordinate amounts of network resources (such as signaling) by these devices and either enable/disable specific services, potentially block misuse, or offer new services to M2M service providers. Some M2M devices may consume smaller amounts of SMS and data, but do not make voice calls (such as meter devices). These devices can be rather costly for operators as there are frequently many M2M devices using signaling simultaneously.
Roaming devices are deployed in a mobile network using either SIM cards purchased by third-parties or other operator SIM cards. Hence, the roaming network is not aware of these devices as they lack a tracking system to effectively trace them. In these cases mobile operators would benefit from a solution to assist them in identifying and managing these roaming devices. To achieve this end, the first step would be to distinguish M2M devices from mobile devices owned by subscribers (phones, tablets, etc.). The second step would be to create a central system where M2M devices can be grouped into categories to better identify their function so the correct charging between the home network and the M2M provider can be applied.
For device discovery, a potential solution would be to collect identifiers (IMSIs, IMEIs) from all operators and equipment vendors, and manage them in a single worldwide database of M2M devices. Operators may identify their M2M equipment in the roaming IR.21 information. Another solution for M2M discovery would be to use big data analytics.
The difference in usage patterns between M2M devices and subscribers’ mobile devices would be easy to analyze. For example, certain M2M devices do not make voice calls. Yet they do send SMSs or data at specific times every day/week/month (for example they report usage on a daily basis at midnight). The amount of data may be fixed in length. Such usage patterns can be defined and matched with the log of events aggregated by the mobile operator.
Signaling optimization for M2M
Another major challenge for mobile operators when dealing with M2M devices is to control signaling usage to ensure the appropriate allocation of resources and QoS for different types of M2M devices. Medical devices for instance, need to perform functions that are critical to the end user. Such devices require a high QoS and priority, and, as a fair amount of these M2M devices send data at a specific time, there is the ever-present risk of peaks and network overload.
A power failure for example, could potentially cause all M2M devices to move to another network which could heavily overload other networks. In addition, the wake-up of a huge number of devices in a network after a power failure recovery could also have a negative impact on the network.
Therefore, the mass of M2M devices needs to be controlled by centralized network congestion and overload management. This functionality can be implemented via policy rules in the LTE network policy and charging rule function (PCRF) as well as 3G networks supporting policy infrastructures, and by other various means for the control of network signaling.
One way of handling such signaling overload is to reject certain devices from being served by the roaming network. The system may determine which devices are to be rejected according to their accumulated profile (for example – malfunctioning devices), and whether to reject or not according to current network load. Other tools for overload management are signaling congestion tools, which can control the operations by queuing and filtering the signaling related to those operations.
Since many M2M devices are inactive for relatively long periods of time (such as a vehicle which sends car maintenance logs every month or so), there is no need to maintain an active record in the network for these devices, but only to manage all the signaling and location changes to reduce signaling pressure on the network. In this case, an M2M profile can be temporarily removed from the roaming network so location changes would not be reported to the home network. When the suspended device initiates activity, the device will re-register to the network automatically. As an alternative, the paging time, which is the time the VLR searches for the device, can be increased to save signaling searches made by the network.
Bandwidth allocated to M2M devices can be controlled either via the profile downloaded to the network components or via a PCRF at the roaming network, which is installed mainly in LTE networks.
A framework for separate M2M billing and charging
M2M applications have different requirements for communications and connectivity compared to human use of mobile network resources. Mobile network operators should be able to manage and control the M2M devices and applications on their networks, and optimize network resources to support M2M services with respect to cost, complexity and performance without degrading the experience for their subscribers.
Business requirements will demand a billing framework to support different charging principles for different M2M devices and applications. For example, the charging principles for a smart metering application may differ from those for an e-book reader, or a glucose level monitoring device. To support such applications, an operator would need visibility of M2M devices, and the ability to distinguish what type of M2M device or application is in use, so it can apply the correct charging principles. In LTE networks, the charging principles would be applied via PCRF.
To sum it all up
M2M’s revolutionary technology has grown in leaps and bounds with the number of currently connected devices in the millions. According to a recent study from Juniper Research, the market for M2M devices is set to reach 400 million units by the end of 2017, with roaming capabilities mandatory for most devices.
M2M will continue to present many challenges for operators in the foreseeable future. Taking control of M2M device activities and effectively detecting roaming devices in the network is first on the list of challenges if operators are to optimize network performance and reduce operational and signaling costs.
