Internet-of-Things: What's so special?
MaharaTech and Google Images

Internet-of-Things: What's so special?

If you haven't already heard of the term IoT, then it is about time! Internet of Things, IoT, is the concept that devices will be able to communicate across the internet, creating a new sphere of internet users, things!

Almost everyone now has a mobile device that needs to be charged daily; what about these 'things' are we going to charge them daily too? What about their connectivity will they have chargeable sim cards? And how on earth are they going to communicate when they are just 'things'.

In this post we will compare typical mobile value chain with typical IoT value chain.

Very Similar!

To over simplify it, they are similar on an abstract level. Both have smart devices (they have processor, memory, and operating system). Both use a wireless network, and connect to gateways to connect to larger infrastructure, and finally reach a aggregator server somewhere which processes and/or stores this data.

Mobile and IoT Value Chains

Major differences!

Generated Data

No alt text provided for this image

Believe it or not, the hype of devices conquering our world is kind of boring! Why? It is because. The majority of data produced, up to 80%, by IoT devices are continuous flow, feed of a single number feed, for example heart rate, glucose rate, machine counts, electricity usage, and so on. The other 20% which requires larger bits of data comes from video captures from boring surveillance cameras or complex ones such as autonomous cars which produces huge amount of data and in various formats, numbers, and video. These data result in "big data", it is easy to see that for self-driving cars, but for an electricity meter producing one number only how do we get to big data? Imagine, this number multiplied by 100 million, and by 365 days in a year or even hourly in the case of a glucose body monitor! Or a factory machine this data is needed every second. Not only is it a storage issue of the data, but image what will this data be like during traffic jams at rush hours, or presidential elections day, or any major event where crowds of requests to the server is condensed? This means that maybe 25% or more of the population is either going or coming back from work at the same time requesting queries of the data aggregated, that is 25 million request on the same server in the same time frame! Least to know that not only is the data you are requesting being requested but every single item on your page app is also being requested, and that is multiple times the load we are trying to imagine! (search request per minutes, rpm for more insights)

Internal and External Networks

No alt text provided for this image

Yes, you are right! IoT devices can connect over wi-fi, over bluetooth and this is what we have all encountered connecting to speakers, printers, smart watches, and TVs over the past years. However the greater challenge is on connecting that shipped parcel across borders, or that oil pipe line across cities, perhaps simply aggregating input from multiple devices for deeper analysis, and decision-making. In brief IoT device, remember 80% of them produce numbers, don't require 5G, and even 4G is over-qualified for those 80% of them. There are special networks that use lower bandwidth such as 2G and 3G, and now some specialized such as sigfox which works in about 90 countries. These have lower bandwidth and so use less battery and can last much longer than the typical couple day charge of our smartphones.

Ok, so that was mainly global networks, or external networks; let us come back to my warehouse, or house, or farm. Do I need 3G in my internal network? Imagine having a mobile device for each sensor in your house, farm, or factory! Of course it is a huge cost just paying those monthly bills! This is where the classic bluetooth, and wifi come to play. But bluetooth only works for about 10 meters, so the more advance are ZigBee which goes as far as 100 meters and LoRa which goes for about 1 km. Further than that then we are going to an external network and SigFox would be a good candidate.

Communication

Ask me, how will they communicate? They are just sensors! Here comes a unique characteristic of IoT devices and that is one of their main benefits; they monitor, track, and they can take action if programmed to do so. Here is the sequence which a typical IoT device will use before taking action.

No alt text provided for this image

Level 1) Monitor actions taken, for example machine start/stop.

Level 2) Send an email alert

Level 3) Send a text message

Level 4) Send a voice message

Level 5) Take preventable action such as shutdown machine before it explodes!

You make the threshold decision of each level of alert to this is typical flow of IoT-to-human communication.

Aggregation Platforms

No alt text provided for this image

These are called Application Enablement Platform. They are another unique characteristic of IoT since really anyone can make an IoT device! These AEP aggregate from many devices (regardless of their manufacturer), they make our life easier; without them we would have to decipher the protocols of each device creating separate codes, this has been done in AEP.

Using these platforms improves data analytics performance, reliability, and enables scalability; without them it is like combining multiple buses together to make a train, rather than make a train from uniform carts.

Sum-Up

And here we are at the end of another blog post, learning something new in the world of IT. In this post we covered 4 major difference between Mobile and IoT value chains.

  1. Generated Data
  2. Internal and External Networks
  3. Communication
  4. Aggregation Platforms

 Drop me a note, reaction, or message if you get this far :)

Sources:

Consultant Bassem Boshra's training IoT Introduction on MaharaTech, available free for Egyptian citizens.

Images either created from notes, taken from lecture, or taken from google, you can use google image to source them.

Bassem Boshra

MasterOfThings, Smart City IoT AEP.

4 年

Good article Ali A.E. A. with a quick good summary as well. Happy that you liked #MasterOfThings #IoT #AEP and SpimeSenseLabs online IoT trainings available on Http://Maharatech.gov.eg Looking forward to see your next article.

要查看或添加评论,请登录

Ali A.的更多文章

  • Lean Happiness Pursuit

    Lean Happiness Pursuit

    I came across a short recipe book on happiness and prosperity by Ken Honda; as I was going through it I saw it as Lean…

  • Got Predictions? What's Next?

    Got Predictions? What's Next?

    For secret sauce of prescriptive analytics aka advanced analytics: patent reference in the end! We have been doing it…

  • Avoid The Suboptimal with Lean AI

    Avoid The Suboptimal with Lean AI

    Future AI I dream of getting rid of all noise information in my day. Imagine if I get the right information.

  • Data Beyond Dreams; The 4 Paths.

    Data Beyond Dreams; The 4 Paths.

    Surely you have data by now, it is never too late to start. Even small family business have a log somewhere that can be…

  • DIY, Deal With Difficult People

    DIY, Deal With Difficult People

    DIY, Do It Yourself, deal with difficult people. Is the divine universe response I got about many people I came across.

  • Make Money from Your Business Data

    Make Money from Your Business Data

    Creating value out of data seems like the natural thing to do in this information age. Data Monetization is how we can…

  • Agile Management - Startups

    Agile Management - Startups

    In this blog we shall explore innovative methods to create startups in an agile way. This exercise can be done once at…

    2 条评论
  • Innovation Menu: Pick and Choose from the Innovation Menu

    Innovation Menu: Pick and Choose from the Innovation Menu

    We are living this reality: continuous updates to our apps, never ending editions of mobiles, technologies (ex: CD/DVD,…

  • Digitally Transformed Business Models

    Digitally Transformed Business Models

    They have changed the way we do business. They are powered with digital technologies.

  • Artificial Neural Networks, are they close to Human Neural Networks? Will they beat us? Or will they complete us?

    Artificial Neural Networks, are they close to Human Neural Networks? Will they beat us? Or will they complete us?

    I heard about artificial intelligence; a lot and for so long! Never have I seen a smarter analogy. I was amused by this…

社区洞察

其他会员也浏览了