GPS devices are NOT CREATED EQUAL!

One of the challenges we face is to explain to customers that in fact not all GPS devices offer equivalent performance despite the supplier quoting that their device contains a GPS module that sounds impressive and seems to have all the right buzz-words and numbers.

Here is an attempt to explain why and what to watch out for…


I am using the term GPS loosely here as the correct term these days is Global Navigation Satellite System (GNSS) – as the modern chips use multiple systems like GLONASS (Russian), Galileo (EU) or Beidou (China), in addition to the US GPS system.


Most systems will work (as in get a position fix) given a clear view of the sky and enough time. But these are ideal conditions, and the real world doesn’t operate that way. Invariably your device is installed on the underside of a trailer, or tucked away under the dashboard of your vehicle, or even inside a container – and we want it to work!

This point is particularly important when it comes to our battery powered tracking devices. A good and smart design can halve the time (and therefore power used) when getting a GPS fix.

POINT 1 – The signals that GPS receivers need to process are extremely weak.

POINT 2 – The GPS receiver needs to go through an acquisition phase where it downloads information about the satellites and their positions before it can use the information contained in the signals to calculate its position. The required signal strength required to download this information (acquisition) is much greater than the signal required for tracking.

POINT 3 – The position accuracy of the fix you get depends on the number of satellites used in the fix and their “dilution of precision” or DOP. If the satellites used to get the fix are close together in the sky then the accuracy of the fix will be less than if the satellites are spread out. In simple terms this translates to “how much of the sky can the GPS see?”.

POINT 4 – the characteristics that a device manufacturer will quote about the GPS module in their product comes from the GPS module supplier and is for ideal lab circumstances for the module, and doesn’t account for how it has been designed into the device. The best module in the world will not work properly with a poor antenna or a noisy power supply.

Things to Look For:

Here are design considerations to take into account when evaluating a product (or designing one):

Antenna Size – for products with an internal ‘patch’ antenna a decent size patch is important. Typically 16mm x 16mm or larger is be a good size.

The CAM-M8 from ublox is a great little module, but take a look at that tiny ceramic chip antenna on there. We used this module in an IoT device that needs a position update every couple of days and is outdoors in a field. In our testing it battled to get a fix when indoors (even next to a window), compared to our vehicle tracking devices with a full patch antenna.

 

Ground Plane Size – the performance of the antenna is generally linked to the size of the ground plane that is placed on. This is one of the reasons why it is harder to get good performance as the size of the device gets smaller.

For example, this is the GPS patch antenna on our Remora device. Nice big patch with large ground plane around it. No noisy components near by. [BTW – the yellow tape on the antenna is to prevent people from shocking the GPS module with ESD (static) when they handle the unit when provisioning].

remora

LNA (Amplifier) – a good design will invariably contain a LNA (low noise amplifier) which as the name suggests can take the signal coming from the GPS patch antenna and boost it.

Location and Orientation – the device design should make it easy to allow you to mount the device so that the GPS antenna has the best view possible of the sky, of course depending on the application. We have customers who install our tracking devices in dashboards as concealed devices (and they work!).

Noise – the PCB design needs to be done very carefully to isolate the sensitive GPS receiver from ‘noise’ on the PCB. This can be in the form of noise coming from the power supply, or from other components (especially digital switching) on the PCB. Try listening to someone whispering on the other side of the room while the TV is turned on at full volume!

Firmware Smarts

In addition to the physical hardware it is critical that a device is *smart* in terms of how the firmware manages the GPS:

Assisted GPS – all modern GPS systems allow the host device to ‘prime’ the GPS engine with information to help it get a fix faster and more accurately. Where possible on our designs we use an offline data file that we download periodically to the device and use that to provide the GPS with Almanac and Ephemeris data.

Smart Backup – the firmware needs to be intelligent in managing when to power the GPS down, and when to keep it in a low power standby mode.

TEST

On one of our recent designs we were testing out different combinations of SAW filter and LNA on the GPS signal. (GPS modules often need a SAW filter to block high power cellular signals if they are placed close to a cellular antenna on a design.) The difference between fairly simple changes on the RF components made a massive difference to the overall performance of the product, like DOUBLE the time to get a GPS fix. This is especially noticeable when the device is in poor signal conditions, like in a forest, urban canyon or even indoors.

So when evaluating a tracking or location device with a GPS in it, please don’t assume that they are all equal in their performance just because the manufacturer recites the GPS module performance characteristics. The best and simplest way to assess this is to do field tests, and put the device through its paces in testing conditions.

 

One thought on “GPS devices are NOT CREATED EQUAL!

  1. Thanks Ken for this insightful post.

    Probably also good to touch on the inherent inaccuracies in modern GPS systems. For example how the US DOD have added inaccuracies into the GPS signals for global commercial use, and not for military purposes.

    One might be using the most advanced GPS/GNSS chipsets, best antenna and design, LNA, and PCB layouts, but you’ll still only get 5m-10m accuracies. Actually fairly similar performance to the cheap GPS designs in smartphones today.

    To negate this, GPS positioning systems, such as in precision agriculture or automated self guidance systems, requiring accurate GPS positioning to cm level need to use Differential GPS positioning (geo-referenced base site to remove the error signals across fleet of GPS receivers within comms range).

    I saw this article recently as well:
    “New GPS algorithms could improve accuracy to centimetre-level”
    http://www.techradar.com/news/world-of-tech/new-gps-algorithms-could-improve-accuracy-to-centimetre-level-1314775

    Wonder how long this will take to work it’s way into the commercial GPS chipsets…. !!

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s