I am exploring some research ideas about architecture for IoT. At present I am come up to a naive idea for IoT architecture, but I am not clear about how to validate the naive architecture for IoT.
Can anyone guide me in this regards, how to validate the architecture for IoT? what tool to be use? or any other guide line?
Thanks in advance
Well, it's hard to say if you keep to generic. Validation as well as used tools depend on what you are proposing.
The validation of the architecture (system architecture) itself is very hard topic. For instance, in couple of our works (it was not only me who was in charge of the works) we used TOGAF (http://www.opengroup.org/togaf/) as the base for the further extension of the criteria catalog. The criteria catalog itself was used to evaluate the architectural decisions and changes in advance. Besides using TOGAF, couple of surveys and interviews with field experts were conducted in order to extend the criteria offered within the TOGAF.
If we will get back to IoT, here situation is even harder, in my opinion, the term IoT is not still well established. I think that a lot of companies, as well as academia scholars, see it from different aspects. This kind of situation add more fuzziness to the overall question you asked.
I hope it helps you.
There is no one architecture for IoT. It depends on the use case, for example industrial control systems have different protocols and requirements than consumer devices. Here is a good blog recently from Joh Mathon on this topic. http://cloudramblings.me/2015/03/04/integrating-iot-devices-the-iot-landscape/
I agree with Trudi. I don't think either IoT-A or iCore can validate their architecture, unless they test it with some real use cases. If you propose an architecture for a specific use case or with a specific feature, you can compare your architecture with already existing ones
Your validation is clearly depends on the use case scenarios....plz mention
Visual basic 2015 integrated with Arduniu UNO or MEga Or mini OR nano
Writing c++ codes for embeded systems of in short micro controllers like ATmega32A which is a digital luxuary
Validation includes the business model as well. The cost, security and integration into existing systems is important. Keep in mind that existing control systems , whether a standard like Zigbee or profibus... or proprietary systems from Honeywell or Siemens or GE will be a design in OEM challenge. If you want to validate a technology I would go to one of the big players or standards bodies and see if you can get their attention. Things like 10X faster, cheaper, more secure will have to be proven.
This landscape of architecture in IOT from device to application.
That largely depend on the protocol and communication infrastructure (m2m, m2c, m2edge).. Next that need to precise part of the question is.. What need to be validated. It's perfectly possible to simulate the data and validate each component separately or build a 'slimmed' proof of concept to validate performance of your architecture, data loss, security, reliability, with scalability in mind. You also need to validate robustness to temporary loss of connection,sensor failure.
- Badges
- Science topic