Calculating the Received RF Noise Power
I have a high-gain receive antenna and I am puzzled because it seems that its power gain is simply ignored in the calculation of the noise power! I thought the antenna gain would affect both the received signal power and the noise power equally.
George Lane: Yes, the gain of your receive antenna is indeed ignored in the calculation of the noise power. The noise power is computed from the CCIR radio noise model where the noise power is assumed to arrive from all directions equally. The use of the theoretical short lossless whip (1/32nd of a wavelength) was to give slightly higher noise for the lower arrival angles. The main assumption here is that any antenna over ground has an average gain factor of 2 since all power directed into the ground plane is reflected back into the upper half space. Thus if the noise arrives from all angles equally, any antenna will produce approximately as much noise power at the receiver as the short lossless whip.
However, if the antenna is loaded and has a resistive loss, then that loss must be removed from the noise power as well as the received signal power. Rhombics and terminated vee antennas are examples where the efficiency factor of the antenna must be subtracted from the CCIR noise power. [Note: the power gain values in the Method 13 and 14 tables include the efficiency loss] Short vertical antennas such as whips and inverted-L antennas also have an efficiency loss which is attributed to resistive losses in the ground plane. You can see these values by running a short fixed length monopole using Method 14. The efficiency values are printed at the bottom of the gain table.
Antennas such as horizontal dipoles, yagi, log periodic antennas are assumed to have no efficiency losses. By the way, this is a bad assumption when the dipole element is closer than 1/16th wavelength over the ground. At that point, ground losses begin to show up in the input impedance of the antenna and are a real resistive loss!
What about the SWWHIP receive antenna that produces less noise power than an isotrope? In the SWWHIP file we can see that an efficiency factor of -4.8 dB has been applied.
George Lane: Basically, you need to integrate the power gain pattern of the antenna in the half space over the ground. This is often called the 'average' gain of the antenna. If the gain factor is 2, then the efficiency is 0 dB. In the case of my shortwave whip model (based on actual measurements of a 15 ft. whip antenna over a ground plane with 15 radials of 15 ft. each at 2 MHz in 1943 at Ohio State University [classified SECRET at the time]), the average gain is 0.331 rather than 2 (10Log(0.331) = -4.8 dB). This value is subtracted from the received noise power, only. The gain pattern of the antenna is applied to the signal power at the appropriate arrival angles.