Phased Array Antennas

Introduction

In antenna theory, a phased array usually refers to an antenna that can be electronically steered.

This is in contrast to a fixed antenna such as a parabolic dish (think DirectTV dish), which has a fixed beam pattern. To steer the beam of a fixed antenna, you would need to physically move the antenna to adjust its pointing. In a phased array, the beam pattern can be steered electronically by adjusting the phase or time delay of the signals fed to each antenna element.

Phased arrays are used in many applications, including radar, satellite communications, and cellular base stations.

Angle of Arrival

Let us start with a 1 dimensional array of N antennas. We'll assume that the antennas are spaced a distance dd apart from each other and the array is oriented along the x-axis.

For this initial exploration, we will consider a receiving array, although the general concepts apply in both directon, as the antenna elements and combiners are generally bi-directional. In practice, antenna elements, time delay blocks, passive combiners, and transmission lines are bi-directional. However, amplifiers and phase shifters are often unidirectional, meaning there will need to be separate feed structures for transmit and receive arrays. For now, we'll focus on the receive array. Although, if it is easier for you to think of this as a transmit array, that is fine too.

The angle of arrival of a wavefront is defined as the angle between the direction of the wavefront and the boresight direction (normal) to the array.

Each of the (initially horizontal) wave front lines arrive one after another.

Note: <OneDimensionalPhasedArrayWaveFronts /> is being converted to this site.

Use the input slider to change θ, the angle of arrival of the wave front, and see how the change in θ change the relative time of arrival of each wave front for each of the antenna elements.

A few notes:

  • Note that we define the boresight direction (θ = 0º) as perpendicular to the face of the antenna. A positive angle θ is defined to the right of boresight, and a negative angle is defined to the left of boresight.
  • The even spacing is arbitrary and is meant to imply (along with the motion) that each wave front arrives at a specific point along the plane of the antenna array a fixed amount of time after the one in front of it. Said another way, the transmitting antenna out of view is continuously transmitting and the speed of propagation is constant.
  • The horizontal line implies that we are in the far field of the transmitting antenna.

Now that we are beginning to understand that the wave front may arrive at each antenna element with different delays, depending on the angle of arrival of the wave front, let's dig in further about what that time difference does to how the signals combine (and what we can do about it)!

Signal combining

As the angle of arrival of the wave fronts deviates from the boresight (θ = 0º) direction, the arrival time delay for each wave front line changes at each element.

One of the main goals of the circuits beyond the antenna elements is to combine (for Receiving arrays) or split (for Transmiting arrays) the signals from/to each array element back to a common point with a common phase for a specific frequency or frequency range (beam squint occurs when steering for specific frequency range but setting delays for a frequency within the range).

In order to achieve this the antenna controller needs to make adjustments to each path to compensate for the arrival time differences to allow the signals to have a common phase at the point of combining/splitting.

Since these antennas are used in communications systems, the common point often is the network modem, with amplifiers and mixers, in addition to the delay blocks, as appropriate for the frequency plan and system requirements assigned to the receiver or transmmitter's portion of the link budget.

Delay Blocks

These paths of combiners or splitters ideally could have a block that compensates for the time delay differences for each antenna element.

These functional blocks are called time delay circuits. They are often implemented with switches to route the signal through or past a set of transmission lines of varying lengths. They are often paired with amplifiers to make up for the signal loss while traversing the transmission lines, depending on the frequency of the system. If you have digital control of the switches and a set of lengths of transmission lines that have a delay of 1, 2, 4, 8, 16, 32 times a base delay value, you can digitally toggle different amounts of delay, often picoseconds or nanoseconds, to each element (or a group of elements).

The above pattern is arbitrarily chosen to be 6-bits. The design of delay blocks can be chosen to match what sets of angle of arrivals need to be compensated for accurately.

In the below example, the red and blue sine waves represent the signals from two antenna elements. The blue sine wave is shifted to the right, indicating that the antenna element it came from received it signal before the antenna element receiving the red signal.

Using the sliders below, adjust the delay of the red and green signals to see when the signals overlap with the blue signal.

Note: <PhaseDelayOfSineWaves /> is being converted to this site.

This illustrates adding time delay to the red and green signals to match the blue signal's arrival time. In practice, the signal amplitudes add together, so aligning the delay of all the signals achieves the desired constructive combination of all the signals.

This example is a simple case of adjusting only a few signals, but in practice, there are many more signals to combine or split in a 1 or 2 dimensional phased array.

Additionally, the delay unit in the example is degrees (of the sine wave, not to be confused with θ, the angle of arrival of the wave front discussed in the section above) to align the visual more clearly to the period of the sine waves**.**

In practice, the delay unit is often time, such as nanoseconds or picoseconds, depending on the frequency of the system.

Phase shifters are often used to emulate time delay blocks, as they are often smaller and easier to fit physically near each antenna element. Phase shifters units are degrees and designed to operate at a specific frequency or frequency range (with some tolerable error for the later case as you deviate from a nominal frequency).

Electronic steering

Let's put it all together.

The concept from the visual applies to the delay blocks for every element in the phased array.

In our wave front example, the delay blocks would be set to match the delay of the wave front arriving at each antenna element, depending on the angle of arrival of the wave front.

Since the arrival time differences vary with θ, angle of arrival of the wavefront, the controller of the phased array needs to compensate (delay in time/phase) for arrival time differences of the wavefront, at each antenna element. Doing so ensures that signals from each element are in phase at the point of combining/splitting for a given θ.

When the signals align in phase at the point of combining, they will constructively combine perfectly.

This intentional adjustment to combine signals constructively at a specific θ is called electronic steering, since the physical plane of the antenna array is not moving, but the signals from a specific θ are combined more effectively than other angles. In practical terms, the electronic steering moves the main beam of the antenna to a specific angle θ. The main beam is the direction of the antenna pattern with the highest gain.

This is the key to the flexibility of phased arrays. The ability to steer the beam of the antenna electronically, without moving the physical antenna.

Mechanical Steering

In contrast, mechanical steering is when someone or something simply moves the antenna. This can be done by hand or by some other method (electronic screws, gimbals, etc.).

Both electronic and mechanical steering attempt to point the main beam of the antenna towards the tranmitter (if the antenna is acting as a receiver) or towards the receiver (if the antenna is acting as a transmitter).

Systems also can employ a combination of electronic and mechanical steering. This is often done to increase the range of angles that can be steered to, while meeting certain performance requirements.

Closing

Common examples would be phased array antennas on the user terminals of satellite internet systems, such as Starlink's user terminals, as well as Satellite's transmitters/receivers, and radar systems.

There are many parameters that differ between electronic and mechanical steered antennas: reliability, cost, beam pointing time, the number of beams supported, the ability to null beams (destructive combining). These parameters are beyond the scope of this introduction, but serve as starting off points for further learning.

Additionaly, it is common for phase shifters to be used instead of time delay blocks. This is often becuase of the availability of phase shifters, in addition to the absence of transmission lines, which are often too long to be placed next to antenna elements. This is especially true when the frequency range of operation is higher, which leads to a smaller distance between antenna elements when targeting an element spacing near a half wavelength (which we have not discussed yet but around 1/2 of a wavelength is a common spacing).

References