INTRODUCTION
1. Navigation is a process involving three steps:
1.1. Establish the track that needs to be followed (i.e. Flight Plan)
1.2. Establish current position relative to the FP
1.3. Execute all necessary guidance actions to correct any deviation in position
2. There are different types of navigation system:
2.1. Pilotage. Using visual ground references
2.2. Astronavigation. Using angular measurements taken between a celestial body (the sun, the moon, a star…) and the visible horizon
2.3. Dead reckoning. Involves the use of visual checkpoints (starting point) along with time, speed and heading measures to estimate the distance travelled
2.4. Inertial navigation. An onboard computer processes speed and attitude, together with information provided by motion sensors (accelerometers, gyroscopes and magnetometers) in order to give the current location from a known start point
2.5. Radio navigation. The application of radio frequencies to determine current position. Aids used include: GNSS, VOR, DME and ADF.
Figure 1. GNSS Antenna
GNSS: GENERAL OVERVIEW
3. Refers to a global system for determining timing, position and velocity using several GNSS subsystems. GNSS is a global term referring to all satellite navigation systems, which include GPS, GLONASS, Beidou and Galileo.
4. GNSS is based on position calculation on the Earth's surface by measuring the pseudodistances from a minimum of three known position satellites. A fourth satellite will allow altitude also to be calculated. Satellite navigation receivers reduce errors by using signal combination from multiple satellites as well as different strategies (eg. Kalman filtering techniques) to merge all the noiseaffected data in order to provide an estimation of position, time (UTC) and speed.
Figure 2. GNSS  Pseudorange corrections
GNSS SYSTEM ELEMENTS
5. Elements.
5.1. Constellation of satellites and onground auxiliary system (maintenance)
5.2. Platform’s GNSS receiver
5.3. Augmentation Systems:
5.3.1. ABAS (Air Based Augmentation Systems) Is an avionics solution that processes the GNSS signal in order to check integrity. It is based on 3rd party onboard devices and special estimation algorithms. The most widely used system is RAIM which uses redundant GNSS signals to ensure integrity and fault detection, giving the ability to warn the user about a defect in a particular satellite's signal. More advanced strategies such as FDE are able to exclude such satellites from the solution.
5.3.2. SBAS (Satellite Based Augmentation System) Is based on GNSS measurements by accuratelylocated reference stations deployed across a specific area. Detected GNSS errors are then transferred to a computing centre and broadcasted using geostationary satellites to complement original GNSS message in the covered area. Systems like EGNOS (Europe), WAAS (USA) and others compensate for certain disadvantages of GNSS in terms of accuracy, integrity, continuity and availability.
Figure 3. SBAS
5.3.3. GBAS (Ground Based Augmentation System) The main difference from SBAS is that the broadcasted signal is sent from ground stations using VHF and UHF bands, being used mainly in airports for traffic control and final approach (LAAS). Systems such as DGPS and RTK are considered a type of Ground Based Augmentation System.
6. UAV Navigation's VECTOR autopilot makes the best possible use of GNSS technology by integrating it within its AHRS.
7. By special request, several GBAS options are also available in order to provide the best possible navigation solution for a particular application.
GNSS: ERROR
8. Global Navigation Satellite System (GNSS) positioning is based on the pseudorange between satellites and receivers.
9. The ‘time of flight’ of radio signals from several satellites to a receiver is used to calculate pseudorange or pseudodistances. The term ‘pseudorange’ is used to distinguish it from true range, as it may be affected by various sources of error in time of flight measurement.
10. Even the smallest timing errors can result in large position errors: for example, a 10 ns timing error might imply a 3m pseudorange error.
11. Various types of error may degrade precision, including the following:

Ionospheric and tropospheric errors

Satellite clock errors

Ephemeris data errors

Receiver quality

Multipath error

Dilution Of Precision (DOP)
Sources Of Errors
12. Ionospheric and Tropospheric Errors. The ionosphere is the layer of the atmosphere between approximately 75 km to 1000 km above the Earth’s surface. This layer contains ions which are electrically charged.
5.1. When the GNSS signal passes through this layer, its interaction with theses ions reduces its speed and therefore introduces an error.
5.2. Ionospheric delay may vary depending on solar activity, the time of year, time of day or location, making it very difficult to predict the induced delay.
5.3. On the other hand, the troposphere is the layer closest to the earth surface. It is approximately 8 and 14 km deep, depending on the location on the Earth’s surface. Tropospheric errors are caused by temperature, density, pressure or humidity changes.
Figure 4. Ionospheric and Tropospheric Errors (see image reference)
13. Satellite Clock Errors. Although GNSS satellites use the most precise atomic clocks featuring nanosecond accuracy, the clock drift phenomena may cause minute inaccuracies which can produce errors that affect positioning.
14. Ephemeris Data Error. These are errors induced by the satellite’s location. An ‘ephemeris error’ describes the difference between the expected and actual orbital position of a GNSS satellite. Because GNSS receivers use the satellite's location in pseudorange calculations, orbital error reduces GNSS accuracy.
14.1. The navigation messages include ephemeris data together with information about the time and status of the entire satellite constellation, called the almanac. Ephemeris are used to calculate the position and the clock deviation of the receiver; the almanac is used to check the visible satellites from a receiver according to the moment of time and its position.
15. Receiver Quality. The hardware used within the receiver may limit precision by introducing inaccuracies in receiver timing.
16. Multipath Errors. Multipath errors appear when a GNSS signal arrives at the receiver GNSS antenna after having been reflected from an object such as the surface of a building (see diagram below). The reflected signal clearly has to travel further to reach the antenna and so it arrives with a slight delay. This delay can cause positional error.
Figure 5. Multipath Errors
17. Dilution Of Precision (DOP). DOP error may be caused by the relative positions in threedimensional space of the satellites used to calculate a position. To get a better understanding, the concept of Geometrical DOP (GDOP) is often used. Poor GDOP values mean ‘bad’ positioning of satellites. On the contrary, ‘well’ distributed satellites produce good values. For further details, please refer to the section "DILUTION OF PRECISION" in this article.
NOTE: This information is displayed by Visionair at the GNSS Status window.
18. Error characterization. Small timing errors may produce large positioning errors. The accumulative effect of GNSS pseudorange error is described by a factor known as UERE (User Equivalent Range Error, sometimes referred to as ‘URA’ in some GNSS documentation). This value expresses the individual error contribution to the global error. The satellite provides a prediction of the maximum total UERE in every message sent.
Source 
Effect (m) 
Signal arrival C/A 
±3 
Signal arrival P(Y) 
±0.3 
Ionospheric effects 
±5 
Ephemeris errors 
±2.5 
Satellite clock errors 
±2 
Multipath distortion 
±1 
Tropospheric effects 
±0.5 
C/A 
±6.7 
P(Y) 
±6.0 
Example:
HDOP = 1.3 (provided by GNSS)
VDOP = 1.8 (provided by GNSS)
UERE = 5.1m (theoretical value)
Standard deviation (RMS error) in 2D Horizontal:
E_2d,rms= HDOP x UERE_RMS = 1.3 * 5.1m = 6.6 m
Standard deviation (RMS error) in 2D Vertical:
E_2d,rms= VDOP x UERE_RMS = 1.8 * 5.1m = 9.2 m
GNSS: DILUTION OF PRECISION (DOP)
19. Dilution of precision (DOP), or Geometrical dilution of precision (GDOP) is a term used in satellite navigation and geomatics engineering to specify the additional multiplicative effect of navigation satellite geometry on positional measurement precision.
20. Understanding navigational DOP values is not simple to explain, but it is possible to establish a simple 2D analogy which makes it easier to understand at a higher level. Let’s imagine a 2D radio positioning system in which a receiver measures the ranges to two terrestrial transmitters to determine its horizontal coordinates. The receiver lies in the intersection of the circular lines of position that are centered on the transmitters.
21. There is some uncertainty, however, in the receiver’s measurements, and so the location of the range circles will be inexact and this will result in an error in the computed position. This error depends on the geometry relating the receiver and transmitters.
Figure 6. DOP
22. In figure 6(a) the transmitters are far apart, this situation gives a relatively small region in which the receiver must lie with some degree of certainty. However in figure 6(b) the transmitters are not as close as in figure 6(a) and so, this gives a considerably larger uncertainty region in which the receiver may lie, We say that the precision in case 1(b) is diluted in comparison to that of (a).
23. The analogy could be that the DOP values depend on high big the uncertainty region in which the receiver may lie, is, and so give an idea of the precision of the measurement that we perceived.
DOP Values
24. Dilution of precision is explained through four or more different parameters. Some of them are related to the other by algebraic expressions. As so, the most important DOP values are the following:
 Horizontal Dilution of Precision. HDOP.
 Vertical Dilution of Precision. VDOP.
 Position Dilution of Precision. PDOP.
 Time Dilution of Precision. TDOP.
25. DOP parameters are nondimensional values and better as they get lower. As a reference a brief table on the meaning of DOP values is attached below:
DOP VALUE 
RATING 
DESCRIPTION 
<1 
Ideal 
Highest possible confidence level. 
1  2 
Excellent 
Accurate enough to meet all but the most sensitive applications. 
2  5 
Good 
Represents the minimum appropriate for making decisions. Reliable in OnRoute navigation. 
5  10 
Moderate 
Positional measurements could be used for calculations, but Fix quality could still be improved. 
10  20 
Fair 
Low confidence level, positional measurements should be discarded or used only to indicate very rough estimation. 
> 20 
Poor 
At this level, measurements are innacurate by as much as 300 meters. Positional measurements should be discarded. 
Position Measurement
26. DOP values highly depend on the amount of available satellites and their position in the celestial dome. As an example of this, lets imagine the following: the tips of the four receiversatellite unit vectors form a tetrahedron (as shown in figure 7). The volume of this geometrical figure is related to the DOP values. The larger the tetrahedron’s volume, the smaller the DOPs. As so, the largest possible tetrahedron for the previous explained 4 satellite model is one for which one satellite is at the zenith and three other satellites are in the horizon at an elevation angle of 19.47º and equally spaced in azimuth; in this special case, for the given number of satellites the resultant GDOP is 1.581.
Figure 7. Tetrahedron formed by four receiversatellite unit vectors
27. However for a GNSS module located on earth’s surface it is not possible to see three belowhorizon satellites, and so in this case the lowest GDOP obtained will be: 1.732, this value is obtained as for one satellite located in the zenith and the other three equally spaced on the horizon.
28. If the number of tracked satellites increases, then the quality precision of the obtained position usually improves, and so the DOP values will decrease giving a better resolution and narrower error, keep in mind that nowadays a normal Ublox M8N DGNSS module has 50 or more channels available for satellite tracking and so DOP values may be highly improved to those of the example above.
GPS WEEK OVERFLOW
29. The Global Positioning System (GPS) system is designed so that from any point on the surface of the planet at least four satellites of the GPS constellation are always visible.
30. The GPS system is based on the known position of these satellites in space in addition to precise timing signals, based on stable atomic clocks, which they transmit to the Earth's surface. The clocks inside typical GPS receivers, however, are less precise and the error that they accumulate, however small, must be corrected in order for them to maintain position accuracy.
31. Each GPS satellite reports information about its position and the precise current time at regular intervals. These signals are transmitted as electromagnetic waves and therefore travel at the speed of light. Position can therefore be calculated by triangulation between at least three satellites and a receiver.
32. Although GPS is mainly used as a positioning system, it may be also used as a precise timing reference for critical systems such us the electrical network, communications and the financial market.
GPS Week RollOver
33. GPS Time (GPST) is determined from a clock ensemble composed of the caesium and rubidium atomic clocks within GPS ground stations and the atomic clocks onboard the GPS satellites.
34. As with UTC, GPST is a weighted mean average time, but with two differences: GPST is available in real time, plus it is a continuous time without leap seconds. It is steered to be as synchronous as possible with UTC. For example, in recent years the deviation from UTC was within about 10 ns. Since January 2009 the difference between UTC and GPST was approximately 15 seconds.
35. This time difference is transmitted in the GPS navigation data message, so that each receiver can calculate UTC.
36. The GPS constellation also transmits the number of weeks which have passed from the initiation of the system to every receiver.
37. The first epoch of GPST started at midnight between Saturday, January 5, and Sunday, January 6, 1980; hence 00:00:00 UTC 06 January 1980.
38. GPST is defined as a week number together with the number of seconds from the start of the week as it is reset every Sunday at 00:00 hours. This allows the correct translation of the date and hour to a friendly format: day, month, year and hour.
39. Since the GPS week is represented by 10 bits in the navigation message, this limits the number of weeks to a range from 0 to 1023, which equals 1024 weeks in total.
40. The first overflow and reset to week 0 therefore took place between 21 and 22 August 1999. The next time the counter will reach week 1023 is on 6 April 2019.
Week beginning at 0000 GPS Time on 
GPS Week Number broadcast by satellites 
08 Aug 1999 
1022 
15 Aug 1999 
1023 
22 Aug 1999 
0 
29 Aug 1999 
1 
24 Mar 2019 
1022 
31 Mar 2019 
1023 
07 Apr 2019 
0 
14 Apr 2019 
1 
NOTE: Ref: U.S. Naval Observatory
41. This phenomenon is known as GPS week rollover and may provoke the failure of inferior equipments which are not adequately prepared.
UAVN's System Solution
42. The GPS receivers used inside UAV Navigation's systems are designed to overcome the GPS week rollover issue. The problem is resolved by assuming that every subsequent week number must be bigger than the week number which is used as the reference.
43. This rollover reference week is coded in the product firmware and is set a number of weeks before shipment of the product and its firmware to a client. In addition, this value can be updated by the autopilot if necessary.
44. The GPS receivers used by UAV Navigation in its products will return the correct date for 20 years (1024 bits / 52 weeks/bits) from compilation of the firmware.
45. The following is an example of how this works in practice:
45.1. Assume that the reference number set in the GPS receiver is 1524.
45.2. As previously explained, the GPS system reset in 1999, so this week was transmitted by the satellites as 500 and corresponds to a week in the year 2009.
45.3. In this case, if the receiver acquires a number of weeks in a range between 500 and 1023, these number will be interpreted as weeks between 1524 to 2047 so the week will be between the years 2009 and first weeks of 2019 (April 2019). If, on the other hand, it receives transmissions with week numbers from 0 to 499, they will be represented as the number of weeks 2047 to 2546 so the week will correspond to a year from 2019 to 2029.
45.4. If the number received is 625, the year represented will be between 2009 and 2019; to be precise, week 21 of the year 2011:
625 = 500 + 125 [range 500 to 1023] 113 = 52 (weeks/year) + 52 (weeks/year) + 21 (weeks/year) 
45.5. However, if the number received is 426, the year will be between 2019 and 2029; to be precise, week 10 of the year 2027:
426 = 52 (weeks/year) x 8 + 10 (weeks/year) 2019 + 8 = 2027 
Figure 8. GPS Week Overflow.
46. UAV Navigation uses a week in the year 2014 as the reference, so the GPS week rollover phenomenon will not cause a problem until 2034 at the earliest.
Summary
47. Implementation of GPS within the UAV Navigation system is robust and welldesigned; it features protection against the GPS week rollover effect.
48. Other, inferior, products from competing manufacturers may not feature such protection and will be prone to failure on 6 April 2019.
49. Protection against this issue has been built into UAV Navigation products up to the year 2034. However, the Company will continue to develop innovative solutions and to provide robust systems which platform manufacturers can rely upon well beyond that date.