Cognitive Computing Assisted IoT

Cognitive Computing Assisted IoT

By Melvin Greer

April 25, 2017


Sensors are integrated in all sorts of things and many of those sensors are communicating with all kinds of machines over the Internet. To that extent, the Internet of Things (IoT) is a reality not a dream. Over the next decade estimates range from 100 million things connected to the Internet up to a trillion things. Devices connected to the IoT will be monitored, analyzed, and acted upon across a number of business verticals. Home devices can provide safety. Connected cars can avoid accidents. Monitored equipment can reveal when parts need to be replaced. An increase in the number of devices making up the Internet of Things (IoT) will also have a significant impact on how the supply chain will operate.

A wave of innovation is transforming all kinds of machines, not just computers, into ‘intelligent’ devices that will reinvent our biggest industries. The Industrial Internet will be one of the most important developments since the Industrial Revolution, giving the world a platform to exchange untold amounts of data, and that will lead to business and technical advantages that have only just begun to be identified. As IoT evolves and expands its reach into virtually every domain, high-speed data processing, analytics and shorter response times will become more important than ever. Fog computing, a decentralized architectural pattern that brings computing resources and application services closer to the edge, along with advances in cloud computing are accelerating IoT deployments.

IoT growth is derived from the connection of physical things to analytics and cognitive computing applications, which can provide insights from device-generated data and enable devices to make “smart” decisions without human intervention. Currently, these technology resources are provided by cloud service providers, where the computation and storage capacity exists.

However, despite its power, the cloud model is not best in environments where operations are time-critical or internet connectivity is poor. This is especially true in scenarios such as telemedicine and patient care, where milliseconds can have fatal consequences. The same can be said about vehicle to vehicle communications, where the prevention of collisions and accidents can’t afford the network latency caused by the roundtrip to the cloud server. IoT sensors are at the edge, but for the now, they lack the computing and storage resources to perform analytics and cognitive computing tasks. Cloud servers have the compute and storage resources, but are too far away to process data and respond in time.

Fog Computing architectures offer an integration of compute, storage and networking resources to mimic cloud capabilities at the edge and support the local ingestion of data and the quick turnaround of results.

A recent study estimates that by 2020, 10 percent of the world’s data will be produced by IoT at the edge. This will further drive the need for more efficient fog computing solutions that provide low network latency together with enterprise class cognitive intelligence.
 
When cognitive computing is applied to the IoT, the result is Cognitive Assisted IoT, can be defined as IoT systems that infuse cognitive intelligence into, and learn from, the physical world. The future of Cognitive Assisted IoT has not been fully realized, however we are already starting to see early generation capabilities. These solutions are replicating some basic functions of human cognition with faster processing speeds to deliver actionable answers to data-based questions. As devices like these become more advanced and more prominent, organizations will be able to drive innovation via Cognitive Assisted IoT in ways that will change the way modern business operates.

The real benefits associated with millions of devices and sensors connected in an IoT world are enhanced by having a learning engine closer to each sensor, displacing any existing rules. Decision-making becomes individual and specific to the sensor or node and purely based on its own experience. In the healthcare use case, health trends and past learning for a specific person is used against a fixed threshold in decision-making. The same idea can be applied across other industries as well.

And because all those devices and sensors are interconnected, their exchange of information and collective learning can offset the significant data and the time required for training and inference while also preparing for the dynamic needs of the solution. In the cyber intelligence use case a particular node exposed to a cyberattack can pass this learned data over the network on the fly, which will help in safeguarding the rest of the nodes. It’s not just the data input that sets Cognitive Assisted IoT apart. In addition to generating answers to numerical problems, cognitive systems can present hypotheses, reasoned arguments and recommendations. They understand an organization's goals, and can integrate and analyze the relevant data to help businesses achieve those goals.

Cognitive Assisted IoT presents as many challenges as it solves. Cognitive Assisted IoT is not a single discipline of computer science. It is the combination of multiple academic fields: from hardware architecture to algorithmic strategy to process design to industry expertise. Cognitive Assisted IoT provides two new characteristics to traditional IoT deployments:

  • Products and services infused with cognitionCognitive Assisted IoT enables the introduction of new classes of products and services that sense, reason, and learn about their users and the world around them. This is the target of Cognitive Assisted IoT, because it allows for continuous improvement and adaptation, and for augmentation of capabilities not previously imagined. This is already happening with cars, medical devices, appliances, and even toys.
  • Enhanced exploration and discoveryUltimately, the most powerful benefit the Cognitive Assisted IoT will deliver is far better insight into an organization's increasingly volatile and complex future. Such insight is becoming more important, as leaders in all industries are compelled to place big bets. By applying cognitive Assisted IoT, organizations can uncover patterns, opportunities, and actionable hypotheses that would be virtually impossible to discover using traditional research or programmable systems alone.

The purpose of IoT is to connect us more closely with the physical world. It provides information with us about the tools we use, the cars we drive, and the cities we live in. But without cognitive computing, the usefulness of this information would be limited by its own scale and complexity. We would only be able to see narrow lanes of insight. The rest would remain in the dark.

That’s why cognitive computing is essential in realizing the true value of IoT.

Cognitive Computing Assisted IoT