When every cm counts, how do I select the right GNSS Antenna?

For applications where every centimetre counts (think UAVs, drones, autonomous vehicles, precision agriculture and the like), a high-precision GNSS (Global Navigation Satellite System) antenna is essential. The steepling growth in these verticals is motivating many vendors to build and supply antennas for them under the premise that it is easy. But selecting the right one is complicated, the selection process may appear intimidating, and picking the wrong one could make the difference between your drone flying in perfect formation…or falling into the river!

Fear not, plucky navigators! This article will guide you through the key parameters to consider when choosing your high-precision GNSS antenna, particularly for deployment in these new verticals.

Start with the vertical (yes, it makes a difference)

With the inordinate number of vendors making antennas for high precision GNSS applications, design engineers are spoilt for choice.  However, and before we even make that choice, let’s consider how innovations in antenna designs are emboldening their effectiveness for some verticals, and enriching many applications with even further improvements in precision and accuracy. While not immediately obvious, debate suggests that a decision here should be made ahead of choosing a vendor (I’ve also written an interesting piece here that describes how the performance of different antenna structures compare).

Ceramic Patch: We should all be very familiar with the architecture of a ceramic patch for GNSS applications. The architecture is mature and well understood, and for most verticals, their performance is very adequate, delivering positional accuracy approaching ~2cm (for a good dual feed, multi-band product).

Helical: While the helical antenna architecture is also well understood, it has often been overlooked for GNSS applications because of an inability for it to be manufactured cost-effectively at scale. With enhanced manufacturing capabilities, the price and performance make it now, well and truly,  a good choice, particularly for verticals like UAVs and drones where weight is a key factor.

Crossed Dipoles: Often the domain of Survey Grade antennas, where accuracy and precision of under 1cm are well documented, a crossed dipole architecture prevails. They too, have proven difficult to manufacture (at scale) cost-effectively. But with some verticals demanding even better precision (and where a Ceramic Patch or Helical just isn’t cutting the mustard), manufacturing innovations have again, paved the way for this architecture to be broadly adopted.

(see my other blog about why the needs of an antenna used in high precision GNSS are different from General Purpose GNSS applications)

The Essential Trio: Gain, PCV, Axial Ratio

With the market vertical understood, and having selected the optimal antenna architecture, our next step is to qualify and short-list our vendors. Assuming we’ve already selected a good multi-band, multi-constellation GNSS receiver, it is important that we select an antenna that matches the receiver for the bands and constellations it supports. Using this as your initial distilling process will rapidly narrow down your list of vendors.

With this step undertaken, we now move to the detailed phase of the selection process. How well a High Precision GNSS antenna performs can be measured with 3 primary parameters: Gain, PCV, and Axial Ratio, and most reputable antenna vendors will disclose them in their datasheets. If they’re not, consider discarding the vendor from your list, as it’s likely they don’t understand how these parameters influence the performance of the antenna for use in a high precision GNSS application (and your drone falls out of the sky!).

Gain: When developing a GNSS based product, as engineers, the first specification we look for is the gain of the antenna element (we’ll not talk about the LNA here), and is normally easy to find. In and of itself, this is a good step, but is improved upon by also understanding the gain of the antenna in all quadrants of the sky. When quoted as a simple figure in a datasheet (at zenith), what is often overlooked is that the gain may degrade by as much as 15dB as a satellite reaches the horizon, so look to assess the antenna for this aspect. As a ranking mechanism, look for an antenna with lower difference in gain rather than just the highest gain at the zenith (that may degrade significantly towards the horizon). Antennas with nominal gain of >1dBic at zenith (where the difference is <10dB at the horizon) are readily available and acceptable, with lower being much better to achieve optimal performance.

PCV: For the surveyors of the world, PCV (Phase Centre Variance) and PCO (Phase Centre Offset) are parameters of an antenna that are well understood. But for most design engineers, even those of us who have designed products with only a modicum of GNSS capabilities, it has not typically been a care factor…for a very good reason. Put simply, a standard GNSS project doesn’t care a whole lot about phase (of the carrier). Conversely, and at a fundamental level, identifying and resolving for the phase of the carriers (of all tracked satellites) is ultimately what will make-or-break a high precision GNSS project. If achieving the highest precision is your objective, a low PCV value is the primary parameter that will deliver it. Don’t be misled though by how this parameter gets specified by some vendors (if indeed it is actually documented). Ensure it is measured across all bands you plan to use (the more the better) and characterised across the entire view of the sky. PCV of ~10mm is typical and acceptable for a good, multi-band antenna (measured from horizon-to-horizon). Be mindful too, that by their very nature, some antenna architectures inherently have a better/lower PCV value, so consider this when choosing.

Axial Ratio: Without getting too heavy into the theory of how it impacts the performance of an antenna, one conclusion we can draw is that Axial Ratio correlates with how the antenna performs in the presence of reflections (often referred to as multi-path). Put simply, an antenna with lower Axial Ratio will perform better at ignoring the reflected signal. From a vendor perspective, many antennas exhibit outstanding Axial Ratio at zenith, often specifying figures <1dB. But consider that many reflections will be incident at or below 45°, making its value at the horizon an important selection criteria.

Now that you’ve short listed your antenna vendors using these three parameters, how you ultimately choose your antenna from here should be straightforward, as your list of options should be very short. And while parameters such as LNA Gain, Noise Figures, Group Delays, Return Loss, and Efficiency could be used to narrow down the list further, you’ll probably find that they won’t, primarily because good values for the three aforementioned parameters typically correlates to good performance in all other parameters. Instead, look to other aspects of the vendor or product, such as IP ratings, mounting method, or even customisation capabilities.

Beyond this, and if you’re still unsure, don't be afraid to consult the experts! The assistance that GNSS antenna manufacturers can provide is invaluable, and they should be able to offer guidance based on your specific application. And if you want objective advise, reach out to SimplexIoT.