Five years ago, sensors began to show up in almost everything—cars, traffic lights, home appliances, medical devices, and more—fundamentally changing the way we performed daily tasks in life and work. The number of these data collecting sensors has grown tremendously, and the true potential is just being realized today in homes, cities, and businesses with machine-to-machine (M2M) communications.
- A Successful Internet Of Things Hinges On M2M
- Use Lua To Develop M2M Embedded Applications
- Understanding The Internet Of Things
For instance, in my own home I built a system to manage my family’s energy use, including how lights are turned on and off and timed bathroom heaters that respond to our daily routines. In my commute each day, I can track the schedules and on-time performance of the ferry I take to work and adjust my departure accordingly. The Message Queuing Telemetry Transport (MQTT) protocol, a standard pioneered by IBM that connects the Internet of Things, makes this all possible.
MQTT Ready To Explode
MQTT is designed for small devices that don’t have much memory or processing power and for low-bandwidth, high-cost, and unreliable networks. It’s designed to minimize network bandwidth and device resource requirements while providing the required assurance of delivery.
As the Internet of Things began to emerge in the late 1990s, I collaborated with Arlen Nipper, the CEO of an American remote telemetry company, to develop the MQTT messaging protocol to make it easier for organizations to quickly and affordably gather, integrate, and make use of sensor data.
MQTT was originally developed to help oil and gas distribution companies monitor pipelines effectively and efficiently. But now, it is being used to help solve the energy crisis and improve agriculture, transportation, and healthcare. It will make getting things done easier, cheaper, and more efficient. By 2020, there could be up to 50 billion sensors, turning the promise of truly Big Data into a reality.
There are many different uses for MQTT including medical devices such as monitors for pacemakers and applications in a new generation of connected cars. While the detail of MQTT is not easily digested by non-technophiles, the key takeaway is that it has the potential of doing for the Internet of Things what the Hypertext Transfer Protocol (HTTP) did for the Internet. It makes it possible for every device on a network to communicate and share information with every other device.
Since we created it, MQTT has been used by companies and whole industries to improve communications between machines—for instance, messaging between sensors on a remote oil pipeline and server computers that monitor the pressure in the pipes and the flow of liquid through them.
OASIS Provides Relief
Recently, the Organization for the Advancement of Structured Information Standards (OASIS) declared that MQTT will be the designated platform for the Internet of Things and is actively working with industry leaders to develop it into an industry standard for organizations to use.
This move paves the way for MQTT to be used widely for applications ranging from power distribution and public safety to retailing, smart phones, and vehicle communication systems. Until now, companies that provide the technology for sensor networks have either used HTTP, which is quite inefficient and verbose, or proprietary technology, which makes it difficult for data from different sources to be integrated.
Take the automobile industry, for example. Every car has sensors that measure gas levels, braking efficiency, and its speed. These sensors have been in cars for decades, collecting information and helping manufacturers build better, more efficient models. But the number of sensors is growing. Blind-spot detection, parking assistance, and lane-crossing alert systems all use sensors to help keep drivers safe.
There is a massive amount of data within the car, and increasingly manufacturers want to study that data. MQTT makes it possible to move this data around efficiently and to make it actionable. The manufacturer can send a message back to the car to alert the driver of a potential problem that may emerge in the near future, spotted by predictive data analytics comparing the car to thousands of others.
Previously, manufacturers have used proprietary systems for message delivery. MQTT will help standardize delivery to help bring uniformity across the industry. For example, Facebook has adopted MQTT for the live notifications it sends to users on iPhones, iPads, and other devices running Apple’s iOS software.
The MQTT platform enables the transformation of a variety of sensor-enabled industries as we see a fusion of cloud, analytics, mobile, and security—the components intricately intertwined to turn data into intelligence and intelligence into actionable decision-making for businesses.
MQTT In Action
Let’s frame this possibility in an example. As you’re driving to the airport, you receive a text message from the airline that your flight has been cancelled. This message sets off an action—the system has automatically rebooked you on the next available flight. But it doesn’t end there. The car rental and hotel you booked at your destination are also notified of your delay. And, for all this “inconvenience,” the airline has automatically notified flight attendants that you should receive free drinks during your flight.
Most people wouldn’t think twice about what makes this amazingly convenient and seamless transaction possible. The reality is that there’s an incredibly complex network of data living in the cloud, where analytics extrapolated the next steps the airline should take and automatically allocated the resources to make them happen.
As the MQTT platform emerges as an industry standard, vendors have begun developing messaging appliances to meet the demanding requirements of an M2M and mobile interaction platform. These products, such as IBM MessageSight, include optimizations for supporting millions of connected devices and accelerating reliable message delivery with high and predictable throughput and predictable latency.
As the infusion of intelligence into the systems around us continues to grow, organizations around the world are beginning to build infrastructures that can collect data about all kinds of things. But the real transformation is how companies can leverage that data to make more intelligent business decisions.
The Internet as we know it is evolving. The hyper-connectivity brought about by the Internet of Things is no longer something that companies can ignore. It’s an exciting time. The world is instrumented with sensors, and we’re just now realizing what we can do with this information and how we can apply analytics to help people make the right decisions for their business. And the MQTT protocol unifies the whole process.
Andy Stanford-Clark is the chief technologist for IBM’s consulting business in energy and utilities for the United Kingdom and Ireland. He is an IBM Distinguished Engineer and Master Inventor with more than 40 patents. He is based at IBM’s Hursley Park laboratories in the U.K. and specializes in remote telemetry, energy monitoring and management, and smart metering and Smart Grid technologies. He has a particular interest in home energy monitoring, home automation, demand-side management, and driving consumer behavior change. He has a BSc in computing and mathematics and a PhD in computer science. He is a visiting professor at the University of Newcastle and a Fellow of the British Computer Society as well.