FROM THE BLOG

Quality Data, Compatibility and Next Steps

Results, reports, graphs, presentations… call them “Deliverables” are only as good as the DATA you collect. Though accuracy is important, what we’re finding out with DataStreams is parameters and consistency are just as important. Questions: What happens when a sensor misreads? What happens if power goes out or the server gets reset? How does your data correct itself and mend while you’re away? These are the current issues that will need to be addressed with our prototype in the future.

So far we’ve created a working prototype system that can be used to monitor and analyse data in our environment – however it’s only a prototype. Engineers and programmers will need to streamline the process, add digital sorting parameters, and create redundancy in the data collection system. This is all part of Phase II of data streams if the prototype is received as viable.

That being said, I’ve made some changes due to hardware compatibility issues that were discovered during our current BETA test series of the modules. xBee modules vary in power and range and though they are intermittently compatible, over a longer period, they begin to cause problems in data transmission. I’ve converted all our radio relays to xBee Series 1 Pro modules that have a theoretical range of around 1 mile. (this distance becomes substantially less once walls, floors and ceilings are in the mix) however, it is a great improvement from our original modules that had a range of 60 feet!

We’re also currently working with Tableau to form some custom coding that will allow us to link data sets from different sensors and share information like Time Stamps and Location to create more accurate reports. We’re having some difficulty getting the programming to where it needs to be, but we’re confident we can have a full working BETA come June.

Thanks to Ryan and all those who have been following and contributing to the conversation. It’s been really exciting to watch this develop.

One comment

  1. Ryan

    on

    This reminded me of a good point to make. All the data that gets collected isn’t necessarily quality data. We can have faults and system errors just like any computer program out there. The key is to be clever enough to spot these differentiations and then know how to interpret them and then correct them.

    Its called, “working out the bugs.” Michael and I are definitely working them out day by day.

Social Media

Stay up-to-date with our latest findings and progress