• 2023-03-07

GPS, GNSS, RTK, PPK – Positioning methods in aerial photogrammetry

GPS, GNSS, RTK, PPK – Positioning methods in aerial photogrammetry

GPS, GNSS, RTK, PPK – Positioning methods in aerial photogrammetry 1024 576 SurveyTransfer

When working with GIS data (e.g., photogrammetry map or model) the surveying instrument is a key factor. The accuracy and precision of location data is also very important through the later analysis. For example, volume calculations between two models can only be successful if the relative positions of the models are sufficiently accurate, with other words, they are referenced. What are the topics of this article? GPS, GNSS, RTK and PPK. Later you’ll see that these aren’t just random letters. 🙂

In this writing we’ll discuss in detail what these positioning technologies mean. By the end of the writing, you will be able to decide which one to choose, as I will list the pros and cons of each technology. 

You should also keep in mind the positioning technology that is optimal and available to you when purchasing a drone for mapping purposes. 🙂


I come across a lot of times, even within the profession, that the terms GPS and GNSS are mixed up. Let’s get things straight! 

The GPS, or Global Positioning System, actually determines our position using only and exclusively the signals from the so-called NAVSTAR (Navigation System with Timing and Ranging) satellites in the United States. 

GNSS, or Global Navigation Satellite System, is the collective name for all navigation satellite families, so it can see more satellites than the subset GPS. Nowadays there are only a few GNSS receivers that are able to receive signals from only one satellite family. It uses different positioning satellites simultaneously to achieve higher accuracy. With most GNSS receivers, you can not only receive signals from the aforementioned NAVSTAR satellites, but also GLONASS in Russia, BeiDou in China and Galileo in Europe. The GNSS receivers provide the horizontal and vertical data referenced to the WGS84 geoid (global reference system).


The measurements can be made with absolute and relative positioning. Using only one receiver, the result of the positioning is the instantaneous value of the three coordinates of the receiving antenna; this method is called absolute positioning. If two receivers are used, the result of the positioning is the difference between the three coordinates of the two receiving antennas. With this we can measure the components of the distance between the two points. One of the receivers operates at a reference point (reference station), while the other one measures at the point that is to be determined. We call this process relative positioning. 

Positioning can be evaluated in real-time and after post-processing. In the case of real-time processing, coordinates can be obtained immediately from the results of the measurements, while in the case of post-processing, the data are processed together after the data collection. 

Depending on whether the receivers are moving while measuring the position, we can talk about static and kinematic measurement. We work with stationary receivers during static measurement. The kinematic measurement is performed with a moving receiver (rover) and a reference receiver (base).


Positioning, in fact, is distance determination. The signal transmitted by the navigation satellites can also be used for time measurement and phase measurement distance determination. The distance measuring satellites transmit two different signals: L1 and L2. The L1 and L2 frequencies are being modulated with two codes for better measurement accuracy.  Thanks to a method called code phase processing, we can expect a measurement noise of approximately 3 meters for the measurement accuracy of better-quality receivers. 

With carrier phase processing a better accuracy can be achieved, compared to code phase processing. With this method the signal evaluation can be done with even millimeter accuracy. In practice the phase processing methods usually use two receivers, so the method can be categorized as relative positioning. In this case we talk about post-processing. The post-processing program calculates the searched coordinate differences from the differences of the carrier phase observations.

Sidenote: The receiver calculates the altitude relative to the WGS84 ellipsoid based on the satellite positions. To get a real altitude, we need to know what is called geoid undulation. This strange word means the distance between an ellipsoid and the geoid. 


Positioning is distance determination. In geodesy (and measurement technology in general) the following errors are distinguished:

• Accidental errors or noise – Noise causes scatter around the actual position. With multiple measurements, the average of the measurements improves the accuracy. They are mainly composed of ~ 1 meter noise of the random code and the internal noise of the receiver.

• Regular or systematic errors – Distorts all measurements in one direction, so increasing the number of measurements will not reduce the value of the distortion on average. The effect of track error, clock error, ionosphere should be listed here.

• Rough errors – Rough errors significantly exceed the measurement accuracy, fortunately do not occur regularly and can be eliminated from the results by increasing the number of measurements.

In real life, not a single error phenomenon can be detected, but a combination of these three error sources.


GNSS receivers are often categorized according to the following applications:

• Navigation (approx. 15-20 m accuracy) – Code phase processing is typically performed. Their advantage is that they are small in size, and the data is immediately available. They do not require office processing.

• Geospatial (Submeter Accuracy) – These receivers are usually ready for carrier phase measurement. They are also compact devices.

• Geodetic (cm accuracy) – These receivers work with carrier phase processing. They can be single or multi-frequency. Some receivers can only detect a single satellite system – although the number is extremely small today – they can usually work with multiple satellite systems. They are equipped with a geodetic antenna with a stable carrier phase center.

• Geodynamic (mm accuracy) – GNSS receivers with a more stable carrier phase center antenna and larger memory. These are long-term receivers and are therefore often powered by an external power source (like solar panels) and some form of wired or wireless communication.


With differential positioning we reduce the effect of regular errors by code phase processing. It can be done with both real time and post-processing. Real-time measurement requires two receivers and a data (radio) connection between them. In this case we place a reference station at a known point which performs code phase processing of the same satellites with a rover. Therefore, this is a relative positioning method. The reference station can calculate the absolute distance of the satellites, based on its own position and the satellite navigation signals. It transmits the differences in distances to the mobile station as a differential correction. The rover corrects the distances measured by itself, based on the corrections received from the base station, thus, the positioning by code phase processing can be further refined. 

In case of post-processing, there is no radio connection between the two receiver stations. The function of the base station is to continuously calculate the differential correction as a function of time and to store the data. The rover station stores the coordinates and data of the satellites used for its calculations as it moves. After the measurement the data from the two stations can be compared according to the measurement times. After that the correction can be calculated by software.


During the real-time kinematic (RTK) positioning procedure, the process is similar to DGPS, but here the reference station even measures phase distances. 

In the case of drones, RTK means that the accuracy of positioning increases significantly, it can even eliminate the need for GCPs. The absolute accuracy could be reduced to cm.

The RTK (as I referred to it) is a relative positioning method, therefore, a reference station and a rover is needed (in case of aerial surveying this is the drone).  Here you have two options. The first is to own a base station and a rover unit. The second is that you have only the rover but not the base station. In this second case the base station will be provided by the CORS (=Continuously Operating Reference Station). The measurement results as well as the coordinates of the base station can be transmitted to the mobile receiver via a real-time communication channel (radio, GSM telephone, mobile internet).

Sidenote: If you decide to use the CORS always make sure you have reference station in your country before spending a lot of money for an expensive RTK drone.

If you use network RTK it’s important to mention the Virtual Reference Station (VRS) method. The essence of this is that after the mobile receiver sends its own position, a central computer generates a virtual reference station data set based on the data of the nearby reference stations and the modeled error effects. This data set means such fictitious results like we’d have placed a reference station near the rover.

Sidenote: In addition to VRS, network RTK can use other methods like FKP (Flächen-Korrektur-Parameter), MAC (Master Auxiliary Concept), PPP (Precise Point Positioning). These are not covered in detail in this article, but if you are more serious about the subject, be sure to look into these as well. 🙂

The advantage of the RTK is the cm accuracy data is available right after the surveying. Due to the high accuracy of the RTK, the deployment and measurement of time-consuming GCPs can be omitted during a drone survey.

However, this system can be really expensive if you’d like to buy your own reference station. If all that matters is that the rover has an RTK receiver and connects to its national reference station, consider that a station near you in the national reference system may break down, thus reducing accuracy (this error can be reduced by using VRS). When working with RTK you don’t have the option to post-correct the results, therefore, the only solution for a survey with poor location data is to repeat the whole survey.


The essence of PPK (post-processed kinematic) is that the rover unit has a GNSS PPK receiver that collects data using satellites and a base station and logs this location data for retrieval after the survey. Just like in case of GNSS RTK technology, you’ll need a base station or a CORS network. Using PPK, when you are done with data collection, you can retrieve the logged location data from a convenient office using separate software, so you can improve the accuracy of GNSS by carrier phase processing afterwards. After the correction, the position data with cm accuracy is available. 

The advantage of PPK compared to RTK is that it does not require simultaneous connection between every unit. In addition, there is no need for mobile internet connection, which can be a huge advantage in many locations (e.g., forest and mining surveys). On the other hand, the disadvantage of PPK compared to RTK is that we don’t immediately get position data, because it requires post-processing. This increases the time it takes to create maps or models. In addition, you need to buy proper software for post-processing, however, this is not a significant amount. Examples for PPK post-processing software are REDtoolbox and KlauPPK.


The question that rightly arises in you is: based on all this, which of the GNSS correction options is recommended?

The question is not so simple, in fact, you’ll need to clarify some questions for yourself first:

  • What kind of areas would you like to survey?
  • What kind of drone you’d like to use and in which constructions (positioning technologies)?
  • How much are you willing to spend (would you buy a reference station or not)?
  • Is the drone supported by PPK post-processing software?

The purpose of this article was to present you positioning technologies (GPS, GNSS, RTK, PPK) and not to decide which technology to use. This is always depending on the circumstances.

Sidenote: If you don’t have the opportunity to use correctional services through a photogrammetry survey and you can only use GNSS receiver, don’t worry! That’s what GCPs (Ground Control Points) are made for. Place them with a determined density and ask a land surveyor to survey the center point of the GCPs. After that you can do the referencing by using the placed points and you can assign the accurate coordinates.  If you’re interested in the placing, measuring and utilizing GCPs I recommend this article.

If you really want to decide between RTK and PPK, don’t miss this comparison article.

Did you like what you read? Do you want to read similar ones?

If you really like what you just read, you can share it with your friends via our social media sites. 🙂