With being sent aiding or correction information, ZED class GNSS modules use data downloaded from the Constellations satellites via the attached antenna to calculate position, velocity & time (PVT). There are inherent errors in the results of those calculations. To an extent, these errors can be mitigated by using correction data.

One example is for a nearby base sration of precisely kown location to send correction data to the target unit. As the base unit’s location is known, it can perform the same algorithm as the target and determine the errors between its true location ans the results of the recent calculation. Those errors can be broadcast from the base unit to one or more targets that share the same ‘sky view’. Typical range for this approach is ~20km at sea level. The accuracy degrades as a function of the distance between the base and targets.

For applications that require accurate correction data for longer variations in distance, there are commercial services that offer local, regional, and national coverage. Please contact us to discuss which services would be worth considering for your applications.

RTK is actually a protocol for the correction data that can be sent from an external agency (to the target unit). The objective is to deliver correction information which can be used in subsequent calculations, in order to reduce the scale of errors in the subsequent computational process.

How do you determine what level of accuracy can be achieved by GNSS?

The usual figure of merit that is used to compare perceived accuracy is called the ‘percentage CEP’.

A CEP50 is used typically, which is a statistical parameter for a circular range about a centre point.

Over a 24 hour period, 50% of the calculated values are expected to fall wihtin the CEP. The smaller the CEP50 value the higher the level of accuracy.

Some aspects to consider include the choice between active or passive. As the name implies, active antenna generally offer superier signal (measured by signal to noise ratio) over passive. Another aspect to note for active antennae compared to theior passive counterparts is that they require a power source which adds to the complexity and weight of the system.

U-blox have created an excellent reference document, which can be accessed here.

No – each unit will determine its own PVT (Position, Velocity and Time) to the best of its ability with the information available to it during the current time, interval or epoch. The overall application will have to assimilate the PVT data of each unit and from that data array, process the relative PVT for all units.

Where there are multiple receivers with a common sky view, they could benefit from receiving (RTK) correction data from a common source. The source could be a dedicated reference or ‘base’ unit or commercial sources e.g. Sapcorda or Hexagon.

In the case where the application uses a Base, the Base unit needs to be ‘surveyed-in’. This typically means rigidly mounted and operated for several hours prior to the intended time for operation. Once the base has been surveyed-in successfully, it will continue to calculate its PVT. As the Base knows its ‘true’ PVT, it can determine errors that occur when performing subsequent calculations. Those errors are the basis of the RTK corrections that can be sent to other receivers in order to further the PVT of those units.

Several factors could contribute to degradation of a GNSS system compared to full potential. These include:

• Optimising the field of sky-view for the antenna. 
• A mismatch of the impedance between the antenna and the RF input of the GNSS model.
• Mitigating the effects of electrical noise in proximity to the antenna, the RF path and the GNSS receiver. Particularly in the frequency bands used by the GNSS systems (and harmonics). Please refer to the relevant Hardware Integration Manual for the corresponding device.
• Careful layout and design of the power supply – especially for the GNSS receiver

Dead-reckoning in this sense is where the GNSS receiver is mounted on a road vehicle. The vehicle can provide additonal information such as wheel ‘ticks’ and direction information to the receiver.

When correctly calibrated, the sensor information can be fused in a DR receiver witht he NSS data. The potential benefits that accrue are that the combination yields more accurate PVT, and that the current PVT can be saved prior to powering the unit down. Assuming that the vehicle has not moved, on subsequent power up the position will be known immediately, without the need for determining the location from unaided GNSS.

U-blox GNSS modules offer two power-saving schemes, ON-OFF & Cyclic tracking. Not forgetting that the user application could enable/disable power to the module.

ON-OFF mode is where the module can be set to be inactive for a long period i.e. intervals longer than 10 seconds between calculations.

Cyclic tracking has the potential to sustatin comparable performance as continuous operation, but yielding up to a 40% reduction in current consumed.

Additional possibilities are for the user application to save remotely the current PVT (once established) then power the module down until another PVT fix is required. In this case the application should allow a time interval of 10s of seconds for the receiver to discern the PVT in the new location.

You can find a useful application note here.