Editing Effective radiated power (section)

== Dipole vs. isotropic radiators ==
Because ERP is calculated as antenna gain (in a given direction) as compared with the maximum directivity of a half-wave [[dipole antenna]], it creates a mathematically virtual effective dipole antenna oriented in the direction of the receiver. In other words, a notional receiver in a given direction from the transmitter would receive the same power if the source were replaced with an ideal dipole oriented with maximum directivity and matched [[Polarization (waves)|polarization]] towards the receiver and with an antenna input power equal to the ERP.  The receiver would not be able to determine a difference.  Maximum directivity of an ideal half-wave dipole is a constant, i.e., {{nobr| 0 dB{{sub|d}} {{=}} 2.15 dB{{sub|i}} .}} Therefore, ERP is always 2.15&nbsp;dB less than EIRP. The ideal dipole antenna could be further replaced by an isotropic radiator (a purely mathematical device which cannot exist in the real world), and the receiver cannot know the difference so long as the input power is increased by 2.15&nbsp;dB.

The distinction between dB{{sub|d}} and dB{{sub|i}} is often left unstated and the reader is sometimes forced to infer which was used. For example, a [[Yagi–Uda antenna]] is constructed from several dipoles arranged at precise intervals to create greater energy focusing (directivity) than a simple dipole. Since it is constructed from dipoles, often its antenna gain is expressed in dB{{sub|d}}, but listed only as dB. This ambiguity is undesirable with respect to engineering specifications. A Yagi–Uda antenna's maximum directivity is {{nobr| 8.77 dB{{sub|d}} {{=}} 10.92 dB{{sub|i}} .}} Its gain necessarily must be less than this by the factor η, which must be negative in units of dB. Neither ERP nor EIRP can be calculated without knowledge of the power accepted by the antenna, i.e., it is not correct to use units of dB{{sub|d}} or dB{{sub|i}} with ERP and EIRP. Let us assume a 100&nbsp;watt (20&nbsp;dB{{sub|W}}) transmitter with losses of 6&nbsp;dB prior to the antenna. ERP < 22.77&nbsp;dB{{sub|W}} and EIRP < 24.92&nbsp;dB{{sub|W}}, both less than ideal by {{mvar| η }} in dB. Assuming that the receiver is in the first side-lobe of the transmitting antenna, and each value is further reduced by 7.2&nbsp;dB, which is the decrease in directivity from the main to side-lobe of a Yagi–Uda. Therefore, anywhere along the side-lobe direction from this transmitter, a blind receiver could not tell the difference if a Yagi–Uda was replaced with either an ideal dipole (oriented towards the receiver) or an isotropic radiator with antenna input power increased by 1.57&nbsp;dB.<ref>{{cite book |last=Cheng |first=David K. |year=1992 |title=Field and Wave Electromagnetics |edition=2nd |publisher=Addison-Wesley |pages=648–650 }}</ref>