Author Options:

HELP: Measuring power density from cellphone tower. Answered

I am working on a project to measure the power density from cell tower. Most projects I saw involve expensive antenna and also a spectrum analyser which is too costly. I came across showing ppl fabricating a target single band microstrip passing through a RF Detector circuit. I think chip like AD8318 could handle it?? But how am I gonna filter if were to use an antenna with range of 200MHz to 2GHz eg. (https://www.sparkfun.com/products/8347) to measure a specific GSM band like 900/1800 MHz from the cell tower? I need help on the filtering and RF detector circuit part!!

The closest resemblance to my proposed system: https://www.favoriot.com/category/how-to/



14 days ago

Not wanting to be un'nice', but this is one of those areas where - when you have to ask questions _here_, you will probably not able to do it.
RF (especially in the GHz range) _IS_ black magic - that's from a radio amateur who designed testers for car phones for a living (by implementing testing equipment with a price tag that could have bought a small plane). Analog was bad - but digital? Urgh....

No, just a receiver (and power measurement) from 200MHz to 2GHz won't help you. There is _A_LOT_ of stuff in between those phone bands going on (e.g. all terrestial television broadcast, avionics, radar ... all depending on where you live).

This "favoriot" project? Yes, definitely something related to waht you wanna do. But orngrimm is very optimistic with his 80%. On the favoriot website, they have error bars of 25dB. Looks good? Well, that's 31622% (for power)! About a factor of 320! And no, that is NOT a (very) bad value for power measurements.

You'd need a least band pass filters for the three phone bands (depending on your area), a power detector circuit for all of them and then there is still the question of what you want to measure? Average emitted power? Peek emitted power? These tower emitt pulses of pulsed data and are quiet in between. So, it is tricky to define, what 'power' is in that case.

Sorry, if that all was somewhat negative. But that is one of those problems that look easy from far away - any cell phone shows you the signal strength with its 5 bars - so measuring that can't be difficult - can it? Yes, it can. The phone _has_ dedicated receiving circuits and 5 bars is more like guessing anyway.


Reply 13 days ago

Hahaha! True! The 80% was pulled out of my a$$ and i wanted only to show that anything close to usable is not acchievable the easy way. But yes, 25dB is horrendous! (My prof once told me, anything above 18dB you better measure with a wet finger. The result is trustworthy the same but you pay less. He was a good chap.).


18 days ago

What you need is a set of very precise and super good band-pass-filters to measure your individual bands. They are very hard to make and sometimes even for me as electronics engineer are borderline black magic.
http://www.avx.com/products/rfmicrowave/filters/ba... could help you a bit. But remember: If you are dealing with such delicate RF-measurements, if you dont want an error or +/-80% you better know your highfrequency routing and all its artefacts and problems.