Antenna models — reading order
- Maxwell’s equations
- Retarded potentials
- Near-field / far-field boundary
- Far-field retarded potentials (this article)
- Array pattern multiplication
- Dipole power gain patterns
Introduction
In the Fraunhofer region the amplitude goes as \(1/r\), and the phase is \(e^{-jkr}\) times a linearised path through the source. The angular part factors into a radiation vector \(\mathbf{F}(\mathbf{k})\)—the object that array theory multiplies.
Motive
We need something an element can contribute as its pattern, and that an array factor can multiply. That something is the radiation vector \(\mathbf{F}\).
Technical development
Under the far-field approximation \(R\approx r-\hat{r}\cdot\mathbf{r}’\) in the phase and \(R\approx r\) in the amplitude,
\[\mathbf{A}\propto\frac{e^{-jkr}}{r}\int \mathbf{J}(\mathbf{r}’)e^{j\mathbf{k}\cdot\mathbf{r}’}\,dV’ = \frac{e^{-jkr}}{r}\mathbf{F}(\mathbf{k}),\]with \(\mathbf{k}=k\hat{r}\). The integral is a spatial Fourier transform of the current—the radiation vector.
How this achieves the motive
Shift a current distribution by \(\mathbf{l}_0\) and \(\mathbf{F}\) picks up a factor \(e^{j\mathbf{k}\cdot\mathbf{l}_0}\). That one property is what produces array factors in the next article.
Sample simulation
import numpy as np
I0_L = 1.0
F = np.array([0.0, 0.0, I0_L]) # z-directed Hertzian moment at the origin
print("F =", F, " |F| =", np.linalg.norm(F))
theta = np.deg2rad(60.0)
k = 2 * np.pi
z_prime = 0.1
phase = np.exp(1j * k * z_prime * np.cos(theta))
print("phase e^{j k z' cos theta} =", phase)
Sample results
The Hertzian radiation vector is angle-independent in this idealisation; the \(\sin\theta\) factor of the electric field appears after the far-field projection. The printed phase factor is the same term that appears when elements are stacked along \(z\).
Next topic
Array pattern multiplication: sum shifted copies of \(\mathbf{F}\) with feed weights.
References
[1] Orfanidis. [2] Balanis.