Editing Radiation pattern (section)

==Proof of reciprocity==
{{Antennas|characteristics}}

For a complete proof, see the [[reciprocity (electromagnetism)]] article. Here, we present a common simple proof limited to the approximation of two antennas separated by a large distance compared to the size of the antenna, in a homogeneous medium. The first antenna is the test antenna whose patterns are to be investigated; this antenna is free to point in any direction. The second antenna is a reference antenna, which points rigidly at the first antenna.

Each antenna is alternately connected to a transmitter having a particular source impedance, and a receiver having the same input impedance (the impedance may differ between the two antennas).

It is assumed that the two antennas are sufficiently far apart that the properties of the transmitting antenna are not affected by the load placed upon it by the receiving antenna. Consequently, the amount of power transferred from the transmitter to the receiver can be expressed as the product of two independent factors; one depending on the directional properties of the transmitting antenna, and the other depending on the directional properties of the receiving antenna.

For the transmitting antenna, by the definition of gain, <math>G</math>, the radiation power density at a distance <math>r</math> from the antenna (i.e. the power passing through unit area) is

:<math>\mathrm{W}(\theta,\Phi) = \frac{\mathrm{G}(\theta,\Phi)}{4 \pi r^{2}} P_{t}</math>.

Here, the angles <math>\theta</math> and <math>\Phi</math> indicate a dependence on direction from the antenna, and <math>P_{t}</math> stands for the power the transmitter would deliver into a matched load. The gain <math>G</math> may be broken down into three factors; the [[antenna gain]] (the directional redistribution of the power), the [[radiation efficiency]] (accounting for ohmic losses in the antenna), and lastly the loss due to mismatch between the antenna and transmitter. Strictly, to include the mismatch, it should be called the '''realized gain''',<ref name="IEEEdict1997"/> but this is not common usage.

For the receiving antenna, the power delivered to the receiver is

:<math>P_{r} = \mathrm{A}(\theta,\Phi) W\,</math>.

Here <math>W</math> is the power density of the incident radiation, and <math>A</math> is the [[antenna aperture]] or effective area of the antenna (the area the antenna would need to occupy in order to intercept the observed captured power). The directional arguments are now relative to the receiving antenna, and again <math>A</math> is taken to include ohmic and mismatch losses.

Putting these expressions together, the power transferred from transmitter to receiver is

:<math>P_{r} = A \frac{G}{4 \pi r^{2}} P_{t}</math>,

where <math>G</math> and <math>A</math> are directionally dependent properties of the transmitting and receiving antennas respectively. For transmission from the reference
antenna (2), to the test antenna (1), that is

:<math>P_{1r} = \mathrm{A_{1}}(\theta,\Phi) \frac{G_{2}}{4 \pi r^{2}} P_{2t}</math>,

and for transmission in the opposite direction

:<math>P_{2r} = A_{2} \frac{\mathrm{G_{1}}(\theta,\Phi)}{4 \pi r^{2}} P_{1t}</math>.

Here, the gain <math>G_{2}</math> and effective area <math>A_{2}</math> of antenna 2 are fixed, because the orientation of this antenna is fixed with respect to the first.

Now for a given disposition of the antennas, the [[reciprocity (electromagnetism)|reciprocity theorem]] requires that the power transfer is equally effective in each direction, i.e.

:<math>\frac{P_{1r}}{P_{2t}} = \frac{P_{2r}}{P_{1t}}</math>,

whence

:<math>\frac{\mathrm{A_{1}}(\theta,\Phi)}{\mathrm{G_{1}}(\theta,\Phi)} = \frac{A_{2}}{G_{2}}</math>.

But the right hand side of this equation is fixed (because the orientation of antenna 2 is fixed), and so

:<math>\frac{\mathrm{A_{1}}(\theta,\Phi)}{\mathrm{G_{1}}(\theta,\Phi)} = \mathrm{constant}</math>,

i.e. the directional dependence of the (receiving) effective aperture and the (transmitting) gain are identical (QED). Furthermore, the constant of proportionality is the same irrespective of the nature of the antenna, and so must be the same for all antennas. Analysis of a particular antenna (such as a [[Hertzian dipole]]), shows that this constant is <math>\frac{\lambda^{2}}{4\pi}</math>, where <math>\lambda</math> is the free-space wavelength. Hence, for any antenna the gain and the effective aperture are related by

:<math>\mathrm{A}(\theta,\Phi) = \frac{\lambda^{2} \mathrm{G}(\theta,\Phi)}{4 \pi}</math>.

Even for a receiving antenna, it is more usual to state the gain than to specify the effective aperture. The power delivered to the receiver is therefore more usually written as

:<math>P_{r} = \frac{\lambda^{2} G_{r} G_{t}}{(4 \pi r)^{2}} P_{t}</math>

(see [[link budget]]). The effective aperture is however of interest for comparison with the actual physical size of the antenna.

===Practical consequences===

* When determining the pattern of a receiving antenna by computer simulation, it is not necessary to perform a calculation for every possible angle of incidence. Instead, the radiation pattern of the antenna is determined by a single simulation, and the receiving pattern inferred by reciprocity.
* When determining the pattern of an [[antenna measurement|antenna by measurement]], the antenna may be either receiving or transmitting, whichever is more convenient.
* For a practical antenna, the side lobe level should be minimum, it is necessary to have the maximum directivity.<ref>{{cite journal|last1=Singh|first1=Urvinder|last2=Salgotra|first2=Rohit|title=Synthesis of linear antenna array using flower pollination algorithm|journal=Neural Computing and Applications|volume=29|issue=2|date=20 July 2016|pages=435–445|doi=10.1007/s00521-016-2457-7|s2cid=22745168}}</ref>