NetObjex Blog.

How IoT compares to the Human Condition

by Raghu B on June 19, 2016

IoT is a revolution currently playing out before our very eyes. Hollywood for a long time has imagined smart machines to have a human like form with a head, torso, and limbs. But let us rethink this rather clichéd image of a smart machine aka robot to what we have currently.

  • Networks: Sensors are plentiful these days and are akin to the senses a human possesses. We have vision sensors, force sensors, motion sensors, sonars for distance sensing, and a whole host of others. So sensors play a role similar to the touch, smell, hearing, sight, and taste senses humans possess.
  • Sensors: Networks are nothing but the channels for bidirectional communication between sensors and the brain. In IoT, sensors communicate with controllers through a variety of protocols. For short distances, we have Bluetooth Low Energy, Bluetooth, Z-Wave, and Zigbee amongst others. Sensors can also communicate with medium range networks such as LoRA, Sigfox, or longer ranges using LTE, Ethernet, and more. Networks are like the Central Nervous Systems that humans possess.
  • Controller: In IoT infrastructures, sensors are connected to a controller. This controller is usually in a localized environment and possesses some intelligence. That could be simple as a Raspberry Pi board running some firmware. In some cases this controller also connects to the cloud for more processing horsepower to do more advanced processing e.g. Big Data Analytics. If we examine cloud based IoT systems, the core logic is divided into 2 parts
    • Big Data: Big Data usually refers to processes that collect data, analyze data and visualize data. Data itself can have 4 orthogonal characteristics:
      • Streaming Data vs Lumpy Data: Basically sensory data usually comes in small bits triggered by some events. In some cases, devices may emit data in a more chunky fashion.
      • Batch vs Real Time Data: Here the focus is on the timeliness of the data to be sent to the Controller for processing. Some data is triggered by events and collected as you go. Streaming data tends to be more real time. Batch data tends to be lumpy in nature.
      • Synchronous vs Asynchronous: This dimension refers to how data payloads are transmitted from the sensors either through some publish/subscribe style messaging mechanism or there is a more traditional web service call to transport the payloads.
      • SQL vs NoSQL: Data structures can differ a lot. Currently, broadly speaking, SQL and NoSQL type data are broad formats. Obviously there are others e.g. File based etc. When data is query-able by SQL then that figures into the type of database we store it in. Likewise, for unstructured NoSQL data.

      In any event, Big Data is akin to the memory that humans possess and the ability to recollect old memories vividly. The visualization of old memories is similar to Visualization Tools e.g. Kibana, Tableau.

    • Rules Engine & Machine Learning: The logical, decision making function is usually governed by a Rules Engine and this can be augmented by Machine Learning. This function decides what do we do with the payload sent by the sensors. There can be many outcomes:
      • Messaging: When a child touches a hot stove, the messaging may be a scream, whereas in an IoT system it may be an SMS or Email to a human that the fire alarm went off.
      • Recording: In a human process, when events occur, we record it either in memory or we write it down somewhere for future reference. For instance, if you receive a water bill in the mail, you may make a note in your To-Do list to pay it. Similarly, when a IoT payload comes in, one may open a ticket in an Enterprise system e.g. Jira as a task to complete.
      • Actuation: Coming back to the example of a child with her hand on a hot stove, the brain sends an immediate command, via the Central Nervous System (Network) to the limb e.g. hand to remove it from the stove. Similarly, in IoT one has actuators (limbs) to take action e.g. turn on Sprinkler if there is a fire.

So, if you compare all of these aspects, one will notice similarities between human actions and processes and with those of IoT devices and ecosystems. The NetObjex platform is a turnkey system that provides all of these functions in a succinct, easy-to-use point & click interface.

Related Articles