Harmonisation between calculated value of the ILS Reference Datum Height and Flight Inspection results
Ivan Ferencz, Head of NAVAIDS division of the Civil Aviation Authority of the Slovak Republic,
M. R. Stefanik Airport, 82305 Bratislava, Slovak Republic
Tel: +421 2 4342 4091, fax: +421 2 4342 4503,
Abstract
Instrument Landing System Reference Datum Height (RDH) is one of the most discussed issue of these days. This Paper describes sources of differences between RDH calculation and measurement.
It is usually highly qualified staff, who are involved into process of the design and optimising of the Glide Slope Mast position. However, sometimes flight inspection finds unacceptable disagreement between expected and measured values of RDH. In addition to, measured values of RDH vary too much [6], assuming that RDH is one of the AIP-published ILS parameters. This is almost true for modern three-dimensional methods of computation of the RDH.
In this article it is shown, that RDH measurement could be improved, when more complex reference figures of the glide path areused as an additional parameters – forward and side slopes of the terrain - will be accommodated directly into flight inspection system software. It is shown, that the uncertainty of the RDH flight inspection results is quite normal and can be expected. The method, how to pass over this uncertainty is proposed.
Introduction
Looking backward, discussions on RDH have begun only a few years ago. The definition of the ILS reference datum listedin the ICAO ANNEX 10 [1] givenas “a point at a specified height located above the intersection of the runway centre line and the threshold and through which the downward extended straight portion of the ILS glide path passes“exists there much longer.
So, why has the problem come out now?
The definition mentioned above requires the view to the glide path from the air down to the ground. It is, of course, not possible without position fixing equipment with three-dimensional (x, y, z) capability.
By using of former position reference systems, as were theodolites or infrared trackers, it was only possible to track an aircraft relative to the head of the theodolite or the infrared tracker. So the view at the aircraft was from the point fixed at the ground to the air. Taking into account guidance for positioning of the tracker stated in former version of the ICAO Doc 8071 [2] (7.4.3.1. ideally the tracker should be sited at the theoretical point of origin of the ILS signals) relative to the Glide Path antenna and not relative to the RWY threshold, it is easy to suppose, that some GP antenna location failures remain hidden for a long time, probably up to now.
Glide Path reference figures
In the process of glide path analysis, the used reference figure has significant influence on obtained results. The GP Error curve can be described as
(1.)
(or vice versa) where
ERR = deviation of DDM=0 from the nominal DDM=0 surface;
EXP = expected GP deviation at aircraft position;
DEV = GP deviation reading at aircraft position.
Obviously, while GP deviation reading is derived from output of the navigation receiver, expected GP deviation depends on applied reference figures of the glide path.
Aiming Point reference
The aiming point reference means that the origin of the reference figure lies close to the intersection of the nominal glide path with the concrete of the runway. Height of the aiming point can be height of the runway threshold or foot of the glide path antenna, or this height can be a result of GP structure analysis. Coordinates of this point in the Cartesian Threshold Coordinate System are [XGP, 0, ZGP]. Expected glide path at vertical plane contained RWY centreline is represented as a straight line.
Aiming point reference matches well ANNEX 10 [1], where an ideal glide path is considered as a plane (Plane of the nominal ILS glide path = Plane perpendicular to the vertical plane of the runway centre line extended and containing the nominal ILS glide path
).
Some flight inspection systems construe aiming point reference as a cone, in this case 0 A surface is
(2.)
where
is promulgated GP angle;
or as a plane with 0A surface
(3.)
Expected GP deviation is deviation of the aircraft at certain point in space from reference 0A surface converted to intended unit, i.e. A. For the case of conic surface is expected deviation
(4.)
Where
DS = displacement sensitivity in A per degree.
Typical feature of the aiming point reference is hyperbolically curved GP error curve.
Picture 1 Aiming point reference GP Error curve - no slope of the terrain
This is in accordance with expectations, because real 0A glide path surface is the conical surface and an intersection of this cone and vertical plane contained runway centreline forms such curve.
When the side slope of the terrain exists, hyperbolical nature of the GP error curve may be accentuated or suppressed.
Picture 2 Aiming point reference GP Error curve - positive side slope of the terrain (0.5)
Picture 3 Aiming point reference GP Error curve - negative side slope of the terrain (-0.5)
Disadvantages of the aiming point reference are:
-Due to curved GP error it is not easy to evaluate the amplitude of GP errors passing ILS “B” point (scaling problem).
-Displacement sensitivity of the reference drift away fromdisplacement sensitivity of the facility approaching the runway threshold due to increasing of the difference between aircraft-aiming point and aircraft-GP antenna distance. This can lead to distinctions in ARDH results and imprecise structure results passing ILS “B” point – details are given in table 1 and picture 4.
Flight inspection systems based on the aiming point reference are capable of direct measuring of the RDH, which is the main advantage of this reference figure.
Table 1 Influence of the tracking angle on GP error, Aiming point reference ( YGP=122m)
Picture 4 Drift of the reference displacement sensitivity demonstrated at three different tracks
GP origin reference
The GP origin reference means that the origin of the reference figure lies close to the foot of the GP antenna. Coordinates of reference point in the Cartesian Threshold coordinate system are [XGP, YGP, ZGP]. Nominal glide path at vertical plane contained RWY centreline is represented as a hyperbola.
GP origin reference does not match ANNEX 10 [1] definition of nominal glide path; on the other hand, well conform to characteristics greater part of the GP facilities.
Touched flight inspection systems interpret GP origin reference as a cone with the 0A surface:
(5.)
and
(6.)
Nominal GP error curve for this reference is a base line, which is comfort for evaluation of the amplitude of GP errors (course structure) over all zones.
Picture 5 GP origin reference GP Error „curve” - no slope of the terrain
When a side slope of the terrain exists, this will be reflected in nominal GP error curve form as reflected at picture 6 and 7.
Picture 6 GP origin reference GP Error curve - positive side slope of the terrain (0.5)
Flight inspection systems based on GP origin reference are not capable of direct measuring of the RDH.
Picture 7 GP origin reference GP Error curve - negative side slope of the terrain (-0.5)
Glidepathoid – the model of the glide path
Obviously, both of reference figures mentioned above are not capable to cover such installations, where terrain is not perfectly horizontal. Often, ILS suppliers must accept some slopes of the terrain within the area preceding glide path antenna – as the forward slope as well as the side slope. People designing layout of such installations really take these slopes into account.
Therefore, implementing their methods into flight inspection systems software can improve flight inspection itself. It should help to find an answer at question, whether the GP system goes such way, as it was designed.
Proposed reference figure fit real glide path installations, because it is possible to adapt the reference figure to conditions at GP antenna site. The used word “Glidepathoid” means the mathematical description of the ILS glide path in space.
A formula (simplified) to describe surface where 0A GP deviation can be expected, is:
(7.)
where
= Electric angle of the glide path relative to the reflection plane in degrees, normally used for GP antenna elements height calculation
= Forward slope of terrain in degrees, positive when terrain climb to threshold
= Side slope of the terrain in degrees, positive when the terrain climb to runway
Note. Formulas use simplification
TAN (+)=TAN ()+TAN () and allowed lateral deviations (y) are within YGP
Expected GP deviation are calculated as deviation from this surface:
(8.)
The nominal GP error curve for this reference is a 0A base line regardless of the slope of terrain, when all parameters are correct.
So, the error curve can be preliminary based on GP antenna site evaluation and finally it will be a result of post processing of GP deviations measured along flight path at points determined by x, y, z values, the task of which is to adjust GP parameters , , , XGP, YGP, ZGP and DS to obtain the best GP error distribution. Any deviation from 0A remaining after this post processing is considered as a course structure.
So the error curve can be preliminary based on GP antenna site evaluation and finally this curve will be a result of post processing of GP deviations measured along flight path at points determined with x, y, z values. Objective if this post processing is to adjust GP parameters , , , XGP, YGP, ZGP and DS to obtain the best GP errors distribution. Any remaining deviation from 0A is considered as a course structure.
Glide Path analysis
Glide Path Parameters
Data required to glide path analysis are:
a)Preliminary values of:
-Glide path origin position, i.e. XGP, YGP, ZGP
-Glide path technical data, i.e. electric angle () and displacement sensitivity (DS)
-Terrain data, i.e. forward slope () and side slopes () of terrain
b)Series of flight data:
-Aircraft glide path antenna positions, i.e. x, y, z
-Measured glide path deviation values, i.e. DEV
For each horizontal position of the aircraft (x, y) is calculated the height of 0A surface, i.e. z0 (x, y) using formula (7) and expected GP deviation (EXP) using formula (8) respectively. From these data are calculated glide path errors using formula (1).
As a next step, an iterative solution is required to find the best distribution of glide path errors. Although any parameter can be iteratively solved, it is recommended to find the glide path origin position by method of geodesic survey first, to eliminate number of variables during iteration process. Further, it is recommended to obtain data from flights below and above glide path to get correct value of displacement sensitivity. So, maximum of variables to be solved during iteration process should be an electric angle of the glide path, terrain slopes, height of GP origin and displacement sensitivity (, , , ZGP and DS).
When all GP parameters, i.e. , , , XGP, YGP, ZGP, DS and course structure are recognized, it is known, how the glide path looks like. What is not known is how such glide path serves to aircraft operation.
GP Angle and Reference Datum Height
From pilot’s point of view, slopes of terrain or GP antenna position are not parameters. Only a GP angle and RDH are published and operationally significant data. Pilots expect, that an electronic glide path will guide them to a point, from which the landing can be safely completed.
Therefore, our interest is pointed up to glide path along localiser (runway) centreline.
To find the value of descent angle and threshold crossing height, it is necessary to make a cut of the z0 surface, which is a result of GP analysing, by the vertical plane contained localiser (runway) centreline. Result of such section will be a curve described with the formula:
(9.)
Then applying of linear regression at the resulting curve between points A and B will result to GP angle and RDH and doing the same between points 1800m and 300m from the runway threshold will result to ARDH value. One of these values can be published as the RDH.
Sources of the RDH uncertainty
Position fixing error
RDH, as the result of GP analysing can be affected by the position fixing inaccuracy. Depending on the basis of used system, there can be an angular or an absolute (x, y, z) error or a combination of thereof.
An absolute inaccuracy of “z” data directly affects RDH by degree of error of measured “z” values. Measured GP angle remains correct. This is possible for GNSS or INS – based position fixing systems. Roots of such error can be, for example, improper correction between the onboard GP antenna position and GNSS antenna or INS centre.
Picture 8 Influence of "z" inaccuracy on GP Error curve
An angular inaccuracy primary affects the measured angle and consequently the RDH.
Table 2 Influence of angular inaccuracy on GP parameters
While the measured GP angle well corresponds to angular error of position fixing system, measured RDH is not so much affected by this kind of inaccuracy.
It is necessary to highlight extreme importance of precise determination of onboard GP antenna along measured flight path.
Receiver centring error
Receiver centring error has a similar consequence as angular error of position fixing equipment.
Table 3 Influence of receiver centring error on GP parameters
Flight technical error
One of features of glidepathoid is, that flight calibration flight profiles are not so demanding on flight technical tolerances. Prerequisite of good repetition of results is using of the correct value of the displacement sensitivity used by glidepathoid.
In Table 4 are tabulated GP parameters for flight with flight technical tolerances 2/3 [4], when displacement sensitivity of the facility differs from displacement sensitivity of the glidepathoid.
Table 4 Influence of displacement sensitivity error for flight within 2/3 standard deviations of the FTT
Reflecting objects
Real environment, into which the glide path antenna is installed, is the most significant contributor of the RDH uncertainty [5]. A degree of the RDH uncertainty depends on the location of reflecting objects relative to the GP antenna, beam bend potential at object location, orientation, size and reflecting factor of objects, speed of the aircraft, used receivers and even on antenna pattern of onboard GP antenna.
Due to wavelength of GP signal (1m), only a small deviation from nominal track can move bends formed by reflecting object along “x” axe and change or even invert their amplitude. Threshold crossing height (TCH) calculated from real zDEV=0 (x) curve differs from flight to flight often up to such degree, that TCH results look totally distrustfully.
Picture 9 Simulation of GP error curve derived from flight at, below and above nominal GP
The obstacle simulated at picture 9 generates quite “good looking” course structure. Nevertheless, while a change in measured angle is within 0.02, excursion in TCH value is 0.5m from nominal value.
Table 5 Change in GP parameters for flight at, below and above nominal GP – calculated from zDEV=0 (x) curves
Application of the method described above helps to get tighter results, as is readable from table 6.
Table 6 Glidepathoid parameters calculated for simulated flight
Example
ILS GP parameters are: = 3.00, XGP=-293.5m, YGP=-120.0m, ZGP=+0.51m, DS=208.3A/, and are unknown.
Picture 10 Real results - Aiming point reference
Picture 11 Real results - Glidepathoid reference before data processing
GP error curves as measured with aiming point reference are at picture 10. Initial GP error curves for the Glidepathoid reference are at picture 11.
Picture 12 Real results - After data processing
Parameters of the Glidepathoid after data processing are: =2.93, +0.08, =-0.43, ZGP remains 0.51m. Resulting GP angle is 3.00, RDH=15.2m and ARDH=15.5m. Obviously, course structure data are more suitable for obstacle situation analysis.
Regulatory aspects
a)Course structure
The application of the course structure tolerances is explained in the ANNEX 10/I, 3.1.5.4: “…bends in the glide path shall not have amplitudes which exceed the following…”. It means, that course structure is determined exclusively by bends only. The course structure is calculated symmetrically from zero GP error line after post processing and tolerance lines have no steps. It is not fully in accordance with the Note 1 as given in ANNEX 10/I 3.1.5.4 because bend amplitudes are not realised on the mean ILS glide path, but it is in harmony with basic idea of the Standard, i.e. to consider a course structure to be a bend.
b)Glide path angle
Calculated glide path angle complies fully with the ANNEX 10/I from straight line which represents mean ILS glide path. This straight line is calculated direct from the model of the ILS glide path ignoring influence of the course structure.
c)Reference Datum Height
The reference datum height is another parameter of the straight line above. The achieved reference datum is calculated similar. Deviations accounted into course structure are again not taken into RDH and ARDH calculation. These deviations can be accounted into TCH and ATCH calculation, which ones then represent threshold crossing height of each individual flight profile.
It is the only way to improve uncertainty of results. Assuming, the RDH is AIP – published parameter, it is mandatory to achieve better uncertainty as is allowed by current ICAO Doc 8071 (0.6m). Otherwise, our flight inspection results look to be not reliable.
The other question is, how serious parameter the RDH is. When RDH is less than 15m, there is a risk, that an aircraft hits the ground in front of the runway concrete. (It looks serious.) When RDH is higher than 18m, it represents a risk, that an aircraft can lose the visual contact with aeronautical lights during an approach. (It is not so demonstrable.) It means that lower of RDH and ARDH should be published as RDH in AIP.
On the other hand, the RDH was more less results of calculation rather than measurement for many years.
Conclusion
The reference datum height depends on GP antenna position. GP antenna position specified by ground engineers is the result of analysis of the terrain and height difference between the runway threshold and the foot of the GP antenna. Hence, ILS GP installation is well described by the same parameters, which were used in proposed model of the glide path.
Applying of the glidepathoid on board or in post flight GP evaluation, it is possible to verify, whether ground engineer has done good job or not.