Why an array is used
A single element has a broad pattern. To achieve more gain, a narrower beam or a beam that can be steered, several elements are combined whose fields add with chosen phases and amplitudes. In some directions they reinforce each other and in others they cancel.
How the pattern is broken down
If the elements are identical and oriented the same way, the total pattern is the product of the pattern of an isolated element and the array factor. The array factor depends only on the positions and the excitations. This separates the design into two questions: which element to use, and how to arrange and feed the elements.
The design decisions
| Decision | Effect |
|---|---|
| Number of elements | More elements give more gain and a narrower beam |
| Spacing | Defines the shape of the pattern and whether grating lobes appear |
| Amplitude weighting | An amplitude that tapers toward the ends lowers the side lobes, at the cost of widening the beam |
| Progressive phase | Steers the beam without moving the array |
| Geometry | Linear, planar or conformal; planar allows steering in two planes |
Spacing and grating lobes
If the elements are spaced too far apart, additional main lobes appear, the grating lobes, which waste power and receive unwanted signals. To avoid them even when steering the beam, the spacing is usually kept around half a wavelength or less; the further the beam is steered to one side, the more restrictive the condition. Very small spacing, on the other hand, increases coupling between elements.
Real effects that complicate the design
- Mutual coupling. The elements interact: the impedance of each one depends on the others and on the angle to which the beam is steered. Mismatch can appear when steering.
- Feed network losses. Dividers, lines and phase shifters dissipate power and reduce the actual gain relative to the theoretical gain.
- Amplitude and phase errors. Tolerances raise the side lobes and fill in the nulls.
- Gain loss when steering. Moving away from broadside, the beam widens and the gain falls.
How it is verified
Each element and the feed network are measured, followed by the pattern of the whole array in an anechoic chamber or by near-field measurement. If the pattern does not match the calculated one, first check the equality of phase and amplitude between channels, and then the coupling. Active arrays are calibrated periodically to compensate for drift. High-power transmission tests must be carried out by qualified personnel, following the applicable regulations.