As M2M systems become more varied and widespread, new networking and platform-building methods can help solve communication, implementation, and scalability challenges.
Independent wireless analyst Berg Insight asserts that the number of cellular network connections worldwide used for Machine to Machine (M2M) communications was 47.7 million in 2008. They forecast that this number will increase to an astounding 187 million by 2014, according to The Global Wiress M2M Market report by Berg Insight.
Why M2M and wireless? Because it simplifies things. Often, M2M involves communications with remote sensors or devices in locations where there is no infrastructure. Wireless M2M provides the ability to monitor remote applications without dealing with or deploying local infrastructure.
I spoke with Dave Richkas, MCU product marketing manager at Microchip, about M2M history, uses, and trends.
Initially, Microchip developed an M2M platform that was GPS/GSM/GPRS based. It was pretty simple to “load and go” – buy a SIM card and deploy it. Richkas mentioned that shortly after the release, AT&T announced the shutdown of the 2G network by the end of 2016 and Verizon guaranteed a 10-year life for CDMA applications. New requests started coming in for CDMA networks since they couldn’t move forward with the GSM-based networks.
Microchip partnered with Twisthink to create the CDMA platform and to achieve cellular network certification. This system used a PIC32 32-bit microcontroller with 512K bytes of flash and 128K bytes of RAM running at 80 MHz, proving that a higher priced processor is not necessary for M2M applications. In addition, they worked with Exosite to provision the platform on the cellular network and enable intuitive user-developed web portals to monitor their M2M applications.
M2M and cellular application examples
Richkas cited an early M2M example involving vending machines:
“An application and early driver of the design involved a bank of vending machines at an airport. One hub was used to collect stocking and health information from the vending machines using ZigBee. The hub then communicates over the cellular network to the centralized system where health and stocking status can be monitored and acted upon.”
I also brought up M2M in a corporate network – for example, perhaps a fitness machine is communicating from a fitness center and logging information for use by doctors and other medical professionals. This intersection of multiple corporate networks and the IT hoops involved can be very challenging. However, offering an M2M solution integrated with the fitness equipment over a wireless network can avoid issues with data traveling on the fitness center corporate network. The data can be uploaded to a portal using the cellular network where the individuals can monitor themselves and track activity levels. If the health club wanted to monitor machine usage levels, hubs could be installed at the fitness centers which would gather the information from the center and relay that information to the cloud where the health club could see consolidated information on equipment usage, diagnostics, and availability for all sites.
More than a platform
Of course, implementing M2M requires more than a platform and there are a number of partners involved beyond just the remote platform itself.
The Microchip MPLAB IDE development suite enables development of the sensor and health application for the embedded device remote platform. Portals are often used as the sites where the remote information is collected, stored, analyzed, and presented to the administrator. In this case, Microchip worked with Exosite on a portal to monitor a spreadsheet snapshot of multiple devices. The portal offers a “drill down” style of information where the top layer provides the overall view, but individual information rows provide additional information about specific remote nodes.
These machines are also sending data across FCC-regulated wireless connections. Getting these systems certified by the FCC and Verizon can be no small task. This is where companies like Twisthink can provide pre-validated systems and manage the certification process to make things simpler and faster for the M2M customer.
Scalability can also be a consideration in M2M. However, scalability has many facets – number of remote nodes, the volume of information transferred within a time period, and perhaps even the variety of I/O and connectivity options.
Richkas mentions that there is plenty of bandwidth available to process many things, from diagnostics to remote control and beyond. The entry point platform prices at around $380, but there is I/O expansion available on the platform itself. Developers can add their own board, access the analog-to-digital converters, and get access to the 5 V onboard power. This platform expansion coupled with the additional portal and interface options provides a complete M2M solution that has room for growth.
The future of M2M
M2M is here and being deployed in a wide variety of systems for an even wider number of uses. Platforms, portals, systems, and analytics must all work together in order to provide robust M2M solutions now and into the future.