M2M solutions follow the same frame-work with regards to data. There are three basic stages that are common to just about every M2M application. However, when it comes to the finer points of machine to machine communication, every deployment is unique.
The process of M2M communication begins with taking data out of a machine so that it can be sent over a network and analysed. The data may be output of a sensor of any type.
There’s a major difference in monitoring an intelligent machine and so-called dumb machine. The intelligence of a monitored machine may be as simple as a temperature sensor, level indicator or contact closure, or it may be an industrial computer system with a Modbus communication port. Monitoring a “dumb” machine may mean directly connecting to and monitoring one or more limit switches, contact closures or analogue outputs. With an intelligent electronic device, it may be possible to simply connect to the equipment’s serial port and ask for the data.
The goal of the M2M hardware is to bridge the intelligence in the machine with the communication network. An intelligent wireless data module is physically integrated with the monitored machine and programmed to understand the machine’s protocol (the way it sends and receives data).
If the monitored machine is configured as an intelligent master device, it may treat the M2M device as a simple wireless modem, loading it up with data and then instructing it to transmit that data to the network. If the machine is just a collection of switches and sensors or is an intelligent slave device, the M2M device can act as the master device. In this mode, it takes charge by periodically polling the device by reading the sensors and switches or by sending data requests through the serial port.
In a high end application like a major electric utility substation, it may be necessary to send a constant stream of real time data describing the machine or process. But in many cases, this is not necessary or worth the cost. In these cases, the M2M device should minimize the amount of data to be sent by constantly reviewing the data, comparing it against programmable alarm limits or set-points, and then only transmitting real time information when a reading is out-of-limits.
In addition the application will typically be programmed to send complete data updates on a time scheduled basis or anytime upon request from the web server.
The data collected has to be transmitted through a communication network. There are several good options for transporting data from the remote equipment to the network operation center. The cellular network, telephone lines, and communication satellites are all common solutions.
The telephone may be the best choice if a line is already installed and the cost can be shared with other uses. Its disadvantage is usually the on-going monthly cost and sometimes the cost and difficulty of installation. Satellite may be the most expensive solution, but is often the best or only solution for monitoring equipment in very remote areas.
The wide spread coverage of the cellular network is the main reason M2M is getting so much attention these days, and it’s usually the method that fits best. There are several methods of sending data over the cellular network. GSM, CDMA and GPRS are widespread and their coverage areas continue to grow. The advantage of these systems is the ability to send large amounts of data frequently. The costs continue to drop.
Connecting to the cellular or satellite network typically requires the use of a gateway. A gateway receives data from the wireless communication network and converts it so that it can be sent to the network operation center, often over the Internet or by a frame relay (phone line) connection. Data security features such as authentication and access control can be managed by the gateway and the application software.
The gateway also has an important role when the flow of data is reversed, going from a network to the machine for data requests and remote control. The gateway still functions as a protocol converter, but this time it takes high-bandwidth Internet protocols and converts them to low-bandwidth wireless protocols so the data is optimized for transfer over a cellular network.
Data from a company’s networked machines usually shows up in one of two places: in enterprise software application the company already uses, or in a standalone system designed specifically for M2M.
Today’s deployments tend to favour standalone systems for applications such as remote monitoring because most M2M application providers specialize in providing these and there can be additional costs involved with integrating new data into existing systems.
Still, the vast majority of opportunities for M2M revolve around taking data out of machines and integrating it with operational data. For example, remote monitoring data can be incorporated into customer relationship management systems for logging service and maintenance history.
Whether the application is standalone or part of a larger system, the common goal is to automate a business process by automating the flow of data to the people and systems that does the analysis. The technology should enable sending the right data to the right place in the right way depending on the circumstances. It should also present data to individual users based on their specific function in the business process.
Of course, none of this technology is specific to M2M; the whole purpose of business software is to keep people from having to do everything manually. The new element that M2M brings to the picture is that now companies have new data to work with, data that is central to the way they operate and the value they provide.
For more information, see: