- Is the Cisco MARS mission going to abort?
- First iPhone worm spreads Rick Astley wallpaper
- 10 stunning 3D buildings made with Google SketchUp
- Open source software ready for big business
- Four reasons to buy (and one reason to avoid) the Droid
Electronic medical-records systems are all the rage in hospitals around the country, promising to improve accuracy and efficiency. EMR fairly screams for a wireless network to support it, however, given that healthcare professionals need access to medical records from pretty much everywhere. Karl Oder, senior director at Rush University Medical Center in Chicago, knows well what it takes to deploy a wireless network that can support not only EMR but RFID and VoIP as well. He and his team, which includes Alden Brugada, manager of telecommunications and networking, tried various iterations of wireless before hitting on their current winning formula.
When did you first get into wireless?
Karl Oder: We had wireless around 1995. It was a very early system that didn't work very well, and we eventually pulled it out. Another
wireless deployment went in around 2000 or so. That was again a proprietary system. It worked better. It was upgraded in 2003
to a nonproprietary Cisco system, and that has since been upgraded to what we have now.
What does the network look like now?
We have roughly 700 LWAPs [Lightweight Access Points], all Cisco [Aironet], models 1231 to 1242. Most of our controllers are
WiSMs [Wireless Services Modules] that sit in the [Cisco Catalyst] 6500 Series chassis. We have six chassis with two WiSMs
per chassis, for redundancy. Each WiSM has two controllers, and each controller can handle 150 access points. So, the capacity
is about 300 access points per WiSM, but we don't go over about 250. We have a [Cisco] Wireless Control System, which basically
monitors the entire environment. The network extends to roughly 10 buildings in a campus environment, plus one hospital in
Oak Park, about 15 miles away. It has its own wireless controller and is connected across T-1s to our main campus.
How has the network evolved since 2003?
The big difference between 2003 and what we have now was the move to LWAPs from the individual access points being managed
one-by-one. We also needed more capacity and more coverage. The biggest impetus for that was an expanded EMR initiative. Initially
we provided only computerized provider-order entry, where the physicians mainly worked on computers in the halls. So, our
2003 deployment only covered the hallways. When you go to a full EMR, with nurses entering vital signs and such, you need
to get the computer inside the patient room. We were also looking forward to deploying VoIP throughout, as well as RFID.
What were some of the challenges in rolling out wireless?
Alden Brugada: Let's start with the 2003 implementation. We didn't do extensive site surveys for that, more like templates. For a 12,000-square-foot
area, we put three access points. The coverage was there, but the signal strength didn't span the floor as we would have liked
it to. For example, we piloted VoIP at that time. It was a passive system, meaning the [IP phone] wasn't scanning for active
access points. You could attach to an access point, walk over to another access point 50 yards away and not attach to that
new access point until you had dropped your first signal. So, we saw a lot of degradation and choppiness.
Comment