Mesh Network Alerts, developed by IBM Research and available soon via The Weather Channel app
Super Bowl LI was a historic day for the sport of football. It was also a historic day for technology. Picture this: wireless and cellular data consumed by fans at Super Bowl games has doubled every year for the last five years. According to Mobile Sports Report, Super Bowl LI broke the single-day wireless data use mark, with at least 37.6 terabytes used. It took almost $100 million in investment from telecommunications providers, thousands of man-hours, and dozens of months to provision the connectivity for the four hour “data feast!” Such growth in data is not limited to sports arenas, it is happening in all walks of life.
In cloud computing — the de-facto computing paradigm today — data must be transported to the cloud for computation to occur. With data growth outpacing bandwidth growth by a factor of two, this paradigm is not sustainable. These fundamental and long-lasting trends are giving rise to the new paradigm of edge computing, an exciting trend in the Internet of Things. The powerful idea behind edge computing is to process the data right where it is generated. A stadium has potentially more raw capacity than some of the most powerful supercomputers in the world when it’s full of twenty thousand people carrying smartphones that sport a powerful CPU, array of sensors, storage and multiple radios for communications – and can be connected to one another. Add tablets, drones, connected vehicles, and other smart devices into the mix, and you have beginnings of the next paradigm of computing.
However, for edge computing to come alive, the devices must be able to communicate without relying on internet or cellular connectivity. A team of IBM Research scientists is creating new peer-to-peer mesh networking technology that allows any modern mobile device to communicate directly with another without needing any Wi-Fi or cellular connectivity. As a first step in piloting this technology, IBM and The Weather Company have announced Mesh Network Alerts for alerting billions of people with potentially life-saving weather information, even when they have no internet connectivity. So how does this technology work? We asked Dr. Nirmit Desai, one of the IBM scientists behind this technology, to provide some details.
Mesh network alerts harness edge computing
How do Mesh Network Alerts work? Does this work on any regular mobile phone?
Nirmit Desai: Yes, this works on most of the modern mobile phones, Android or iOS, without needing any special jailbreaking or hardware extensions. All modern mobile devices come with Bluetooth and Wi-Fi radios and our software uses these radios in innovative ways to discover and connect with devices around you.
How close do these devices need to be?
ND: Interestingly, this varies a lot across device models, but typically a pair of devices within a few hundred feet of each other can communicate. However, messages can hop over multiple intermediate devices to reach devices much farther away from the original sender. Of course, unlike the current internet, the messages will take longer to reach and the amount of data we can transport is still limited. This is why the weather alerts are a great first application for this.
What are some of the challenges you faced?
ND: Mesh networking is a widely understood concept in the scientific community. There are routing algorithms, performance analysis techniques, and some small-scale field tests to learn from. However, what is new here and probably the most challenging still is enabling this on millions of regular mobile devices without any special software or hardware.
Battery is a scarce resource and device radios consume a lot of power. Deciding when and how often a device should use the radios for discovery has a major impact on battery life. We had to come up with computationally simple yet effective algorithms to bring down battery consumption to less than one percent per hour on most devices.
Another major challenge has been to ensure that a mesh network can be formed even when the devices are in our pockets. This is essential, because an alert may arrive at any time and, on receiving it, each device must forward the alert to neighboring device to ensure that everyone gets the message in time. In this way, mesh networking exemplifies the power of communities. However, this is very difficult to achieve as the mobile operating systems such as iOS and Android are designed to support interactive apps where the users are actively engaged.
Can anyone send information on such networks?
ND: We can protect both the confidentiality and the integrity of the messages. In doing so, we are bringing together IBM’s security expertise and The Weather Company’s experience in delivering highly usable apps to millions of users. For the weather alerts scenario, it is essential that hackers cannot spread misinformation, e.g., fake alerts, but the content of the messages is not confidential. We employ state of the art cryptography techniques to ensure that users are only shown alerts that verifiably come from The Weather Company.
Are there applications of this technology beyond weather alerts?
ND: Most certainly, we are just getting started! Networks become congested or unavailable in crowded areas, e.g., sports arenas, music concerts, theme parks, cruise ships, disaster zones. On the other hand, mesh networks thrive with such high density of devices. We may see many of the social-sharing apps leverage mesh networking to carry data. Gaming apps have already applied this among a restricted group of players.
In a new paper published in Nature Electronics, IBM researchers demonstrate the smallest ever built DRAM memory cell, fifty years after its invention. The new DRAM cells feature potentially low power consumption and an unprecedented small footprint. They could be therefore particularly appealing for implementation in mobile devices or as cache memory.
For the second year in a row, researchers at the IBM-Illinois Center for Cognitive Computing Systems Research(C3SR) won a competition challenging experts worldwide to design low-power embedded systems for Internet-of-Things (IoT) applications. The 2019 Design Automation Conference (DAC) System Design Contest’s objective: create algorithms that can accurately detect and locate objects from images taken by […]