It is not 10 db of signal loss it is 10 db of return loss, or RL, which the article is talking about.
Return loss is the portion of a transmitted signal that cannot be absorbed by the load (the antenna in this case) and is reflected back to the source (transmitter).
In this article 90% of the transmitted power is absorbed or transmitted by the antenna and a small 10% of the total power is reflected back to the source (the transmitter) due to the mismatch from a theoretical perfect 50+J0 load impedance.
I won't get into waves reflecting from antenna back to the source and then back again to the antenna and then back to the source but suffice to say this is what happens in a mismatched system.
RL is calculated as follows RL=20log((ZLoad-Zsystem)/(Zload+Zsystem)) where Zsystem is typically 50 ohms, the result is displayed as negative db numbers.
Think of it in a way as to how far the load impedance is off of a theoretical/perfect 50+J0 ohms of impedance.
With a perfect 50+J0 Z load you would then experience maximum power transfer where the source Z matches the Load Z and all the transmitted power is radiated by the antenna.
In a typical perfectly matched antenna case the return loss would be an infinite - number but typically a RL figure of -30 db to -50 db is what a high acuracy power meter or network analyzer would provide for a system in near perfect impedance match.
I say near perfect because in reality you will never experience a true 50+J0 load impedance.
I am studying M Sc in elecrtrical engineering at BTH in Sweden. My course works have been completed and now I am trying to prepare my thesis proposal on Antenna technology. These information helps me a lot to think about my future research area in Antenna.Thanks. If possible then please send me the current research topics in this area which will help me lot to select my research topic.
Blog Doing Math in FPGAs Tom Burke 15 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...