@Anand.Yaligar, things are not that easy to automate or connect as they look on paper. For expample the biggest challenge for the smart devices to become integral part of our life is their seemigly lack of intelligence on handling unknown scenarios.
@DrQuine, i too am skeptic about IoT at the current stage of development and wonder if we cannot smartly integrate our personal electronic devices then how easy/difficult would it be to connect the seemingly different devices.
The Internet is well established, what the Internet of Things needs is reliability and usefulness. When solid applications are identified thaqt can be implemented through IoT, they'll take off. We're seeing hints with cars that automatically unlock when we touch the handle (if we're carrying the key). When doors open and lights seamlessly turn on when we enter a room below a certain light level, the IoT will start to realize the long sought potential. Reliability is a key ingredient. Sitting still reading a book does not mean that I want the lights to go out and devices that fail due to dead batteries decrease the quality of life they are supposed to improve. It often seems on-line that as many things are broken as work; we need a much higher quality standard for IoT to provide a net benefit.
@Jonas: Yes, we can start with home appliances which can directly be used by connecting the power source to the IOT device inside the home, we can also start with automobiles and things inside the framework of industrial internet. Anything is possible.
"With the huge demand for lithium coming from electrfying our vehicle fleet and other demands of the pwoer sector, i don't think the IOT would pose such a big problem.Anyway , the forecast is for 4X growth in lithium demand by 2020, and i haven't seen(in my short search), any mention of problem to supply such demands.
As for trillion sensors - if that's true , we'e talking about a multi-trillion dollar opportunity - supplying that is a very nice problem to have. "
Sensors are the fundamental units of an IOT and making a trillion sensors would need supply chains to start coping up with things like transportation, management, vendor acceptance etc.
With the huge demand for lithium coming from electrfying our vehicle fleet and other demands of the pwoer sector, i don't think the IOT would pose such a big problem.Anyway , the forecast is for 4X growth in lithium demand by 2020, and i haven't seen(in my short search), any mention of problem to supply such demands.
As for trillion sensors - if that's true , we'e talking about a multi-trillion dollar opportunity - supplying that is a very nice problem to have.
Bert, samsung seems to have opened it's API for appliances, so i'm not sure other companies will have much choice - because being forced to buy everything from the same company is quite a big issue for people going to buy some appliance or device.
In my personal opinion not all IoT devices have to be wireless and rely in batteries. At least in process plants many devices are connected to two-wire networks that provide both DC power and superimposed communication. These networks can be intrinsically safe, non-incendive, or installed using explosion proof hardware so they can be used in all kinds of explosive atmospheres such as in refineries and petrochemical plants. FOUNDATION fieldbus is a good example of such a protocol for industrial use. Similar solutions could be used in homes as well. Cars and other motor vehicles use CAN bus on board. However, having said that, there are many cases where wireless sensors is the only practical solution.
In my personal oppinion, not all devices need to have an IP address. They just need a unique identifier and the ability to communicate. A USB webcam is a good example. It works well without an IP address. What is important is that the small devices are networked to a linking device that has an IP address. The applicaiton protocol on the small local network will be the same as over the IP network - just different media for the local network.
Applicaiton protocols are crucial. Every automation area will need an application protocol tailored for its application. Process automation has FOUNDATION fieldbus and WirelessHART. Building automation will have another. Motion control a third. Home automation will have its protocol. One size does not fit all. Without standard application protocols there will be no interoperability to make devices work together. Moreover, we also need profiles to enable interchangeability - i.e. replacing devices with another brand when it fails. I have seen this in process control. Profiles are asbolutely neccessary to enable replacement of failed devices without having to reprogram or reconfigure controllers.
Lastly you need device descriptions like Electronic Device Description Language (EDDL) for process control devices in order to access all features in all devices. A standard way to access non-standard features.
A Book For All Reasons Bernard Cole1 Comment Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...