5G EMF measurements the easy way
Well we've said 5G just because it's the buzz terminology and your internet search might find us a little more easily but what we really mean is field strength measurements around mMIMO & beam forming antennas (as well as the other telecoms and broadcast emitters you might find on a rooftop). Measuring 5G signals is tricky because most likely it's a very wideband signal and portable spectrum analysers at a reasonable cost don't have sufficient resolution bandwidth (RBW) to look at the signal all in one go and then of course if your measurements are for ICNIRP compliance you need to use an analyser coupled to an isotropic probe. Then because of your limited RBW it's not necessarily straightforward how you capture a maximum value...
Or alternatively you can use a broadband meter and probe which quickly gives you a result for all frequencies present. That said the type of broadband equipment needs a little bit of thought. With a 100kHz to 6GHz diode based broadband probe we recently measured a maximum field strength of 91 V/m in a public area with the main contributions coming from 4G (700 MHz) & 5G (3.5 GHz) signals. For those of us that know our ICNIRP general public reference levels (yes we should get out more) we know we've got a problem, 61 V/m is the highest value listed. So on the face of it we've got a non-compliance. The next step is of course to take a time averaged measurement but before we get to that how about taking a look at the specification of broadband diode probes. They all have a relatively small 'square law' or true RMS region. This means if you've got a couple of signals (or more) that take you out of this true response region you get multiple signal error and it can be sizeable. Use a shaped probe (this just means the output is shaped or weighted to the ICNIRP reference levels as they change with frequency) like the Narda ED5091 and this error is much reduced plus it automatically gives you a result in % of the ICNIRP levels (so you don't have to remember the V/m values). Sound easy, yes it is - to avoid costly false non-compliances you just need to start with the right probe, a shaped one.