A modern solution for multi-aperture and phased-array antenna characterization

Tue, 05/07/2013 - 1:03pm
Alex Dickson and John S Hansen, Agilent Technologies

Forming a beam or multiple beams with an electronically scanned array antenna has found utility in a number of applications. These include radar, communication systems, radio astronomy and others. An array antenna consists of multiple antennas or radiating elements that each transmits a signal where the relative phase of these signals is manipulated to form a beam of specific shape and direction. Combining array antennas with beam-forming techniques enhances spatial selectivity and interference avoidance. It is no surprise that array antennas have been replacing mechanically steered antennas for several decades.

Initially, passive electronically scanned arrays (PESA) were introduced where the signal is generated at single frequencies by a small number of transmitters and a phase shifter is incorporated on each radiating element of the array. This enables the antenna to have greater beam agility and allows multiple functions in a single antenna. The advantage of active electronically scanned arrays (AESA) is that a separate module is mounted directly on or very near the surface of the antenna for each element in the array unlike the shared transceiver with PESA. Each transmit/receive modules (TRM) can generate and radiate its own independent signal of different phase and frequency as needed. This provides the ability to produce sub-beams effectively enabling the array to operate as several antennas at once, each operating on a different frequency. Beam agility derived from the flexibility of this architecture allows for frequency hopping and beam scan patterns that are not predictable (good for avoiding jamming).

Solid state amplifiers have advanced to the point where TR modules can be light and compact while transmitting the high power needed for longer range radars. Gallium Nitride (GaN) is an enabling technology due to its power-handling and thermal capabilities.  The conversions between analog and digital domains (DAC/ADC) will soon be located directly within the TR module itself (see figure 1) putting the digital signal right next to the antenna, simplifying the overall design of the system.

There are many challenges associated with the test and calibration of phased array antennas. The more elements that make up the array the longer the time it takes to fully characterize the antenna. In a phased array antenna with hundreds or thousands of elements, where it is necessary to characterize each element in a relative way to the others, the ability to accelerate the test by using multiple coherent measurement channels is a significant benefit.

The element-to-element phase and magnitude (gain) errors of the various components in the array antenna are significant limitations to its overall performance. Since phase is used to steer the beam in a phased array, the errors introduced by the misalignment of the radiating elements must be calibrated out so that the antenna operates efficiently and accurately. Below we will focus on the static phase and gain errors across elements and describe new methods for producing a set of calibration measurement data to correct for these errors.

Depending on the signal, there are two possible methods to analyze the cross-channel response by measuring relative phase and gain.

The first method is a narrowband approach that uses a swept or stepped tone and a narrowband receiver to measure one frequency at a time and perform cross-channel computations in the time domain. However, this method is limited to narrowband measurements.

The second method uses a broadband stimulus and a wideband receiver to measure all frequencies simultaneously and compute the cross-channel spectrum. The ideal measurement solution has the flexibility to use both methods.  A wideband digitizer with Digital DownConversion (DDC)provides this flexibility and is a unique solution because of its adjustable bandwidth.  Let’s look more closely at DDC and its benefits for this application.

As shown in figure 2, a hardware-based DDC is a two-stage digital signal processing block that processes data taken directly from the analog to digital converter (ADC) at full sample rate. Then, after frequency translation and decimation, the data is stored as complex I&Q samples to the digitizer’s memory. A DDC can also be created in software, but the consequence is that software DDCs run much slower and rely on the data at full ADC sample rate to be first off-loaded from the digitizer for processing.

So what exactly does the DDC do for you? As part of a digitizer it allows you to isolate the signal of interest then improve the SNR and dynamic range within the bandwidth of the signal of interest by reducing the amount of integrated noise. Additionally, it extends the amount of signal capture memory or reduces the amount of data that needs to be transferred for a given duration. Since there is less data to analyze, a DDC can reduce the workload of the post processing algorithms.

The way the DDC improves sensitivity for phase and amplitude measurements is by reducing noise in the time domain. The noise density remains the same, but less noise gets integrated into the measurement as the span is reduced. You can see the depiction of multiple waveforms in figure 3 and how, at a given threshold, we can visualize the phase differences between these waveforms.  Determining the actual time that a noisy waveform crosses through the threshold is what can be difficult. You can see from the three plots how it is more precise to determine where the waveform crosses the threshold with reduced noise on the waveform.

Figure 4 shows a generic block diagram of a test system used in the test and calibration of array antennas. The challenge as already stated is to measure relative phase and amplitude of radiating elements in a phased array. It is often desirable to test multiple pairs of elements at a time to accelerate test speeds when dealing with large arrays containing hundreds or thousands of elements.  The signal path is from left to right, originating with the antenna array under test. Then the signal travels through a path of several stages of signal conditioning and down conversion. The goal is to take the RF/µW signal from the antenna elements and frequency translate it down to an IF that is within the BW of the digitizer. To maximize the use of dynamic range of the digitizer, it is typical to amplify or attenuate the signal to levels that fall close to the full scale range of the digitizer being used. In some cases a low-pass filter is also employed for image rejection.

The digitizer is the back-end of the signal measurement chain. When testing a phased array antenna configuration, it is often desirable to test multiple pairs of elements in parallel to accelerate test speeds. Therefore, a multichannel digitizer with phase coherent inputs (< 1 degree phase difference) is required. In a phased array antenna with hundreds or thousands of elements, where it is necessary to characterize each element in a relative way to the others, the ability to accelerate the test by using multiple input channels is a significant benefit. As technology evolves and antenna configurations continue to have higher and higher densities, a scalable platform that can accommodate additional channels in the future becomes equally important.

In addition, a digitizer with sufficient 3dB analog bandwidth is also necessary in order to use it to characterize signals across all of the different functions possible in the phased array. Modern active electronically steered array (AESA) antennas do not only transmit and receive continuous wave (CW) tones, but also often have signals of bandwidth as in the case of communications or different types of modulation. For example, there are several radar configurations that use pulsed RF or Barker codes, or other forms of modulation, that increase the amount of bandwidth that is used. Therefore, a digitizer that has enough bandwidth to encompass a variety of high bandwidth test conditions is also required.


Share this Story

You may login with either your assigned username or your e-mail address.
The password field is case sensitive.