Chapter 7: Seeing the Invisible: The Site Assessment
5. Finish setting up the spectrum analyzer by doing the following
• Set the span to 60 MHz. This setting ensures that the analyzer duly notes any AEN on either side of the 915 MHz center frequency. You want to know if any signals are even close because a device broad- casting at 901 MHz can cause interference.
• Set the resolution bandwidth to 100 kHz to ensure that you record reasonable levels of interference.
• Set the video bandwidth to 30 kHz to obtain a smooth plot on the spectrum analyzer.
Figure 7-2:
An antenna at the center of a target area.
125
Chapter 7: Seeing the Invisible: The Site Assessment
• Set the amplitude attenuation to 0 dB so that the spectrum ana- lyzer displays a discernible signal-to-noise level (making it easier to see the interfering signals above the noise floor).
• Turn on maximum hold so that you can capture the energy of every channel in the band you are testing.
After the antenna is in the middle of the proposed interrogation zone and attached to your correctly tuned spectrum analyzer, you should see the virtual screen on your laptop, or the video screen on the spectrum analyzer should be active. If you see either one of these, you’re ready to start the testing.
Measuring for AEN during Normal Operations (And Beyond)
The Full Faraday Cycle Analysis is the foundation for building an RFID network, and the goal of that analysis is to have a perfect foundation for the structures being put on top of it. For a full analysis, you need to do the following:
Identify all the ambient electromagnetic noise within the facility.
Log data over the course of a full business cycle (all the shifts) to under- stand any changes that happen at different times of the day.
Measure specifically at each interrogation zone to correctly assess the strength of signals relative to where you’ll install readers.
Rove around the facility to make sure you looked in every nook and cranny for rogue AEN.
Triangulate any sources of interference while roving about the facility (in other words, get closer and closer until you find the source of the interfering AEN).
Run all the possible machinery and equipment that is likely to make electronic noise in the interrogation zone while recording the data to make sure no potential source is overlooked.
Address the potential interference found by either eliminating it (for example, upgrading a wireless barcode system from 915 MHz to 2.4 GHz or finding a creative workaround (such as deploying only hand-held RFID readers that do not broadcast at the full power of a fixed location reader and don’t create as much interference).
Map out the interrogation zones on computer-aided drafting (CAD) drawings or blueprints and make sure that those zones are noise-free.
With your equipment set up, you’re ready to begin. The following sections explain in more detail how you accomplish each part of the analysis.
Testing key points around the warehouse
As you test, you need to log the data:
If you are using a laptop to log the data or employing a software-based spectrum analyzer (like a National Instruments model),you should set your virtual monitor to record information every hour and actively log it to the hard drive.
If you are using a spectrum analyzer without a laptop,you should come back and take a digital picture every two hours over the course of the normal business cycle. After taking the picture, clear the video dis- play by resetting the video screen and begin collecting data again.
This initial measurement process is effective, but only represents one data point in a facility. How large the warehouse is and how strong interfering sig- nals are will determine what you pick up from that one location. So if you have a large warehouse, you have to set up the same test procedure at sev- eral locations within the warehouse to increase the accuracy.
You want to make sure you test for AEN close to all the potential RFID reader interrogation zones (usually the dock doors, conveyor or sort stations, and so on).
If you do notice any significant spikes on the monitor, you’ve caught some interference. It is important to note the location, the time, and the frequency of the interference and try to map a pattern (for instance, does the spike occur every hour when the security guard makes his rounds?) or try to narrow the time period in the next day or two to figure out the source of interference (did a FedEx truck arrive to pick up packages during the window when you had interference?).
One drawback to this static testing methodology is that it is difficult to find the location of any interference. That’s why you follow up with other tests, which I explain in the next two sections.
I’ve been a wild rover for many’s a year
The next step toward increasing the accuracy of your Full Faraday Cycle Analysis is to take a roving data capture of AEN. This test is particularly effec- tive if you share a facility with other tenants who may be running systems
127
Chapter 7: Seeing the Invisible: The Site Assessment
that are separated from yours by only sheetrock walls. To take a roving cap- ture, you need the following:
A portable uninterruptible power supply (UPS) or similar battery back- up device. You need the ability to power the spectrum analyzer for 20–30 minutes. These are available from American Power Conversion for under $200 at most computer stores.
A golf cart, shopping cart, or similar means of wheeling your equipment around a warehouse safely.
A willing friend to act as a human tripod and carry the antenna around next to the golf cart.
Blueprints or CAD drawings of the facility come in handy, too.
To do the roving capture, follow these steps: