The Rise and the Research of Machine-to-Machine (M2M) Communication

This is a research blog dedicated the communication technologies that are related to Machine-to-Machine (M2M) communication. We will primarily provide information related to our research project MassM2M, but also analysis of the relevant literature, technology trends, and research ideas.

What is M2M?

Probably today you have used M2M communication if you paid by a credit card. The terminal (a machine) in the shop connects to a server (another machine) in order to approve the transaction. M2M is about communication between devices, objects, things, which is different from Human-to-Machine (H2M) (e. g. “Googling”) or Human-to-Human (H2H) (e. g. “Skyping”).

Techno-economical forecasts indicate that in the coming years M2M communication will become massive, connecting tens of billions devices. Wireless chips have grown in capability and power efficiency, while shrinking in size and cost. It becomes affordable to embed wireless chips in many diverse objects and make them “digitally visible”, similar to the way a person is digitally visible through Facebook.

M2M becomes massive also in terms of diversity across applications. Wireless M2M networks are instrumental to manage the complexity of tracking, fleet, and asset management. The industrial sector can widely apply M2M in monitoring and control of processes and equipment. Radio Frequency Identification (RFID) in the retail industry is an example where M2M enables real-time visibility of the individual items. The M2M showcase is the smart grid: the evolved power grid where a rich information flow is used to balance the electricity production (e. g. windmills), distribution, storage, and consumption (e. g. large industrial capacities).

M2M has a large transformational power to make the processes more efficient by saving time, costs, and energy. But the present optimism about M2M is also fueled by its promise for new business models and the large innovation potential. The companies that use M2M, such as the industrial sector, can introduce novel features in their products and deploy new, information-intensive services. The companies that provide M2M services, such as the telecom operators, see them as important alternatives to the flat-data-rate-like services.

Skeptics may question the upcoming M2M revolution, pointing, for example, that RFID has been around for years, but failing to become massive. This is true; but it was also true that the phrase “smart phone” had been around from 1992, while it lifted off with the iPhone in 2007. Recall that the mobile phone started with voice as a single application at its focus, only to become our assistant and chief entertainer. M2M communication has started slowly with a wide range of applications, so our expectations should be very high for the upcoming billion connected devices.

Research on M2M

M2M services are already up and running in the networks of many mobile operators. Furthermore, there is are very active standardization processes related to M2M in different bodies, such as ETSI or 3GPP. So, is there a need/space for carrying out fundamental research on M2M communication technologies?

Our answer is (clearly) yes. Specifically, there are two (expected or predicted?) features of M2M communication that allow one to pose new and interesting research questions:

  • The “Massive” feature: Technology predictions say that by 2020 there will be 50 billion connected wireless devices, spanning a wide application range: smart grid, metering, control/monitoring of homes and industry, e-health, etc. Wireless networks need a revolutionary reengineering to be able to embrace the massive number of devices. In many cases M2M traffic will feature short data packets, where the useful data is comparable in size to the signaling overhead used to send that packet. To- day the networks can efficiently carry large data from few devices; the problem is how to carry few bytes from a large set of devices (machines). In a nutshell, sending 100000 bytes from one device is very different form sending 1 byte from 100000 different devices; the latter will clearly consume much more resources for signalling/coordination.
    •    The requirement for dependability. M2M communication becomes vital for various control, monitoring, and industrial processes, where it is critical to keep the wireless link alive during 99.99+% of the time. This is in a stark contrast to many existing systems, such as WiFi, which works fine around 95% of the time, but offers zero data rate under harsh receiving conditions. Increased dependability means that the wireless link is available almost all the time and, under harsh conditions, it can scale down the data rate in order to maintain reliable connection.

Other M2M issues that pose interesting research questions are security and device management. They are only peripherally related to our research, but it has to be noted that they also essential for wide adoption of wireless M2M deployments.

Regarding our research approach, we have two different tracks. In one track we investigate protocols and algorithms for rather generic communication systems. An example is Frameless ALOHA, where we are exploring a new concept for massive random access. In the second track we adapt our research context to a particular system, such as LTE, see the article on Code-Expanded Access in LTE. Although it may seem far from the cutting-edge research, we are very interested in the GSM system. We believe that the GSM networks should not be put out of use, but rather to be re-engineered and dedicated to M2M traffic. Even the research on GSM can become fundamental if we understand the GSM protocols and structure as design constraints for new protocols and signalling schemes that should be built on top of it. Our initial activities related to dependable wireless communication have been chiefly related to the university spinoff Wisecan.