Getting ready to The Internet of ThingsPosted: May 1, 2013
“If real is what you can feel, smell, taste and see, then ‘real’ is simply electrical signals interpreted by your brain”. Morpheus, The Matrix.
Internet of What
Nowadays hottest technology trends in IT and computing are big data, mobility, artificial intelligence, wearable computing, and apparently Internet of Things. Those are buzzwords. What is IoT (Internet of Things) anyway? Is it about more devices with internet access? Smart networks? Smart homes? The Matrix?
In few words Internet of Things is the global network connecting any smart object. It is all about huge network of interconnected devices and sensors, their interaction. M2M (machine to machine) communication is very related term and is about underlying technologies for communication between devices utilyzing RFID, sensor networks, NFC, Bluetooth, Zigbee, WiFi and WiMAX etc. The ultimate difference from Internet-as-we-know-it is that nearby 50-petabytes of data currently available in Internet is almost all captured and entered here by human. Internet of Things data comes from Sensors.
Some stats. Now (as of beginning 2013) about 2.2 billion of people use internet services. Analysts expect from 20 to 50 billion of interconnected devices by nearest 5 to 8 years (and only 5 billion of them are expected to be smartphones). What are others? ~1 Billion vehicles (smart transport), ~3 Billion utilities, the rest is smar homes and smart industries.
When these devices are connected in smart way, it looks like “The Machine” is here. Assume billions of sensors and servers are the single brain. There is great video on TED about 5000 days of internet.
For humanity, the network of sensors promise bright opportunities, like smarter homes, smarter bodies (with additional wearable computing like pebble or nike plus), smarter interaction with environment and just-in-time context-based information (smart glasses)
For enterprises, IoT provides ability to control and improve manufacturing process, better automation, better analytics.
From us, engineers and technology people and companies, IoT trend will require new skills, new technologies and new solutions.
Connecting the dots
The main technology challenges for IoT include collection, management, analysis of huge amount of information, sensors, and networks, which traditional computing system are not able to solve.
Assume the Machine is like the single human. There are main stages in information flow from outside world through our sensors to brain.
1. Sensor understands external signal (hand on hot plate)
2. Sensor sends signal through the internal neural network to main or peripheral brain
3. Brain processes signal and sends signal to act (muscle get hand from hot plate)
Likewise, The Machine demands are:
1. Sensors. From human muscle information like MYO to manufacturing machines sensors – they provide continuous flow of data (not information yet) about some attributes of monitored “thing”. Modern sensors measure temperature, blood pressure, heart activity, muscle activity, radiation, movement, precise location and hundreds of other parameters. As of 2013 there are dozens of start-ups which are developing new sensors, with adding some infrastructure besides sensor alone – connecting it to smartphones and backend (another sample is lapka)
2. Channels. This is about M2M communication, new protocols, IPv6 adoption, Bluetooth, NFC etc. Trends there are higher-speed (more data), wireless (less cords), standardized (vendor independence).
3. Processing. As alternative to traditional OLTP and OLAP applications, new high-load systems are empowered with distributed big-data processing technologies, in-memory databases, smarter algorithms (e.g. growing use of video sensors raised new wave in evolution of image recognition technologies).
What else? The latest, but very important piece is analytics and presentation of information (visual, audio, video) for humans. Since we do not expect The Machine to live alone, and want to use it for sake of smarter humanity, important decision will still be made by people. These decision will be supported by analytics – useful information, extracted from huge amount of data. Technology is emerging – look to MessageSight from IBM.
The big picture:
And according to technology roadmap by SRI, we are still on the long road:
How we are getting prepared
As I am currently working in R&D (research and development) at SoftServe, we are aware of new trends and do researches in following directions:
* M2M communication
* Big Data, including social analytics, big images processing and big data visualization
* Future UX (user experience) and HCI (Human-Computer Interaction)
* Machine Learning (and particular application of it to Image Recognition)
* Mobility and sensor networks