Fog Can Help Shape the Future of IoT

Whenever I hear about a serious train accident, mugging or shootout on the streets of a city, my thoughts often turn to Fog Computing.  The same is true when I too am stuck idling in a traffic jam or at home and there’s a power outage during a winter storm or a summer heat wave.

Why do I think about Fog Computing?  Well, my job at Cisco is to not only identify the latest disruptive Internet of Things (IoT) technologies, but also to validate where they might be applied to improve overall quality of life..  Whether it’s drones, artificial intelligence or robotics, my passion is to accelerate the art of the possible.

 

Consider Fog Computing.  Fog extends cloud computing to the edge of the network.  This provides a virtualized platform for compute, storage and network services between devices and data storage centers in the cloud.  Because of its low latency, location awareness, real-time interactions and wide geo distribution, Fog Computing can sense and respond to situations in the real physical world almost instantly.

The speed and power of Fog to connect people, data, processes and things opens up a new world of practical solutions.  For example, Fog Computing, when combined with sensors and wireless networks, can immediately alert the train operator as soon as there is trouble on the tracks, such as a slow-walking pedestrian or a stalled vehicle.   With Fog, energy loads can be automatically re-balanced or re-routed to alternative sources during spikes in demand or low availability.

In a Smart+Connected Community, acoustic sensors deployed around streets that are connected to Fog Computing infrastructure can identify gunshots, perpetrators, victims, accidents, or even cries for help with high accuracy while also alerting appropriate authorities.

When realizing these scenarios, my thoughts also often turn back to Cisco Live! in San Diego. A couple of months ago, our host team (Carlos Morales, Pedro Silva, Rui Nogueira and myself) had the extreme pleasure of jointly demonstrating these ideas with technology partners from Cisco and other companies.   In a highly collaborative functional demo, Cisco and Intel created a Fog-enabled Smart City using Lego® blocks, Arduino boards, Raspberry Pi computers, Zumo Robot Cars and Intel Core i7-5650U Fanless Computer as the Fog Node controller.

The purpose of the demo was to bring awareness to the key business drivers for Fog Computing – namely latency, bandwidth and location awareness. As an example, the demo contained a pedestrian detection system implemented on Fog Node using Raspberry Pi Camera Module and OpenCV Computer Vision software. With a simple button press, visitors could simulate an operating mode as if the software was running in the cloud instead of local Fog node. Higher latency and longer turnaround times to and from the cloud were suicidal for the pedestrian almost every single time. Fog Computing provided a much more reliable and practical paradigm to implement a Smart Traffic Light System.

The demo also captured an emergency evacuation use case. With the press of another button, users could trigger an emergency in the Lego city. Smart street lights became even smarter in providing unblocked intersections to emergency vehicles and in directing public flow to safe evacuation zones. Low latency and precise location awareness were the winning components of the Fog Computing system in this scenario.

Real-life deployments of Fog Computing in Smart Cities – not to mention factories, healthcare, energy and elsewhere — could drive many other outcomes that were beyond the scope of this demo. The enormous social, economic and environmental potential of Fog Computing, when linked with the Internet of Everything, is somewhere between the art of the possible and the imagination.

Stay tuned for more developments happening around Cisco IoT System and Fog Computing.