EP1678459B1

EP1678459B1 - Method and device for the reduction of influence of the multipath effect on the own position measurement of a beam guided object and the rf beam guided object control system using it

Info

Publication number: EP1678459B1
Application number: EP04791201A
Authority: EP
Inventors: Herman THALES INTELL. PROPERTY Benthem de Grave
Original assignee: Thales Nederland BV
Current assignee: Thales Nederland BV
Priority date: 2003-10-14
Filing date: 2004-10-12
Publication date: 2011-10-12
Anticipated expiration: 2024-10-12
Also published as: NL1024532C2; ES2372819T3; WO2005038386A1; EP1678459A1; ATE528608T1

Abstract

This invention relates to a method and the corresponding device for the reduction of the influence of the multipath effect on the own position measurement of a beam guided object and the RF beam guided object control means using it. In guided ammunition control, the use of RF beam guidance induces that the own position measurements of the guided ammunition with respect to the guidance beam are affected by multipath effects in the case of low flying targets. An object of this invention is a method for the reduction of the multipath effect on the own position measurement of a beam guided object comprising the limitation of the variation of the measured position into a predetermined interval [F2, F1].

Description

This invention relates to a method and the corresponding device for the reduction of the multipath effect on the own position measurement of a Radio Frequency (RF) beam guided object and the RF beam guided object control system using it.
In RF beam guidance, the beam guided object has to follow a RF beam aimed in the desired direction (beam riding). In flight, the guided object measures its own position with respect to the RF beam and translates these measurements into appropriate commands for its own control means. The use of RF beam guidance induces that the own position measurements are affected by multipath effects when the object is at low altitude above the sea.
Electro Magnetic (EM) energy propagates as waves in the atmosphere and when it hits a surface, reflection occurs. In the case of an object at low altitude above the sea, a relatively large part of the EM energy arriving at the object will come from reflections at the sea surface. The smoother the sea as compared to the RF wavelength, the more energy will be reflected. In smooth seas, the reflected energy has a strong average value (specular reflection); in less smooth seas a random effect (diffuse reflection) will dominate.
Due to differences in the path lengths of the direct (R₁) and reflected (R₂) waves, they will arrive with a (very small) time difference at the object. As a result, this time difference translates into a phase difference. If the phase difference is small, then the two will add up to a larger value than from the direct path only, a phase difference near 180 degree can lead to almost extinction. Whereas the range difference will vary with target distance, this interference or multipath effect will also show a range dependent fluctuation. The smaller the wavelength (higher radio frequency RF), the higher the frequency at which these fluctuations will occur: more specular reflection peaks and lows within a particular range bracket.
To enable the guided object to determine its own position with respect to the main direction (boresight), a number of predetermined beam position offsets can be introduced, either simultaneously as in a monopulse system or sequentially. Modulated by the beam pattern, the different beams will give different responses that the guided object uses to estimate its own position.
The multipath effect influences the accuracy of the estimation process. In particular, this is due to the fact that the indirect paths will have different angles (and gains) with respect to the boresight at each beam position. Thereby, the multipath effect introduces additional errors in the own position measurement, even when the object is at the boresight.
In the case the object is a guided ammunition, the RF guidance beam is aimed by means of a guidance beam antenna in the direction of the target to be intercepted. In this case, multipath effects will occur when engaging low flying targets such as Seaskimmers.
The multipath effects directly will influence the fly-out of the guided ammunition and reduce the probability that the guided ammunition will hit the target. Even when equipped with a proximity fuse that will detect the target when passing and detonate the guided ammunition's warhead, the effectiveness of the blast and fragmentation on the target will be less with a larger miss-distance (lower kill probability) with respect to the target.
Thus, the multipath effects degrade the operational performance of the RF guided ammunition in terms of kill probability and keep-out range.
A method to reduce the effects of multipath is the use of RF agility, i.e. to make use of different frequencies in subsequent measurements. Because the positions of the multipath peaks vary in range with the radio frequency RF, some frequencies will be affected less than others at a particular range. The guided ammunition processing can make use of this by e.g. selecting only measurements that have a sufficiently high received Signal-to-Noise-Ratio (SNR) : multipath peaks are associated with low SNR values. The larger the bandwidth (bandwidths of >10% of the main operating frequency are preferred in this respect), the more effective the use of RF agility will be.
State-of-the Art transmitters using Travelling Wave Tubes (TWT) are designed to operate with optimum performance in a relatively narrow band around the main operating frequency. Larger or smaller radio frequencies RF are possible (within a bandwidth of about 10% of the main operating frequency), however, at the cost of reduced output power. To a certain extent, these effects can be compensated for, however, with cost consequences.
Another method to reduce multipath effects on RF guided ammunition control could be by means of measurement data filtering. RF beam guidance essentially involves the closed loop control of the guided ammunition. Closed loop, in this case means that the guidance influences the position of the guided ammunition with respect to the guidance beam, which obviously directly influences the input to the guidance, i.e. the measured own position. Filtering basically delays the influence of the measurements taken into account in the guidance. As an example consider the following case:
Due to e.g. a target manoeuvre, which can be followed almost directly by the guidance beam antenna, the guided ammunition builds up an offset in position with respect to the guidance beam antenna bore sight. As a result of the filtering delay, this offset initially is not noticed. When the guided ammunition eventually has noticed the offset and starts its correction manoeuvre, this again is not noticed until after the delay. Eventually when the guidance already has corrected for the offset successfully, still an error is noticed and the guided ammunition correction command is maintained, which will result in an overshoot of the desired position, etc.
The effect of filtering is a less damped guided ammunition motion (stronger and longer lasting oscillations), which eventually will result in larger miss-distances with respect to the target (=less kill performance). Note in this respect that the guided ammunition, as any dynamic system, is a sort of filter itself already. If the filtering on the measurement data is too strong, actually the complete system can become unstable and the guided ammunition will never arrive at the target.
Thus, high bandwidth RF guided ammunition control requires a more expensive transmitter and measurement filtering degrades the stability of the guided ammunition.
The patent US 3,290,599 A discloses a power modulator for transmitter beam scan. A drawback of such a modulator is that it is affected by multipath effect.
This invention solves the above-mentioned drawbacks by limitation of the guidance beam error measurements to legal errors, i.e. errors due to target or guided ammunition manoeuvres.
An object of this invention is a method for the reduction of the multipath effect according to claim 1.
Further features and advantages of the invention will be apparent from the following description of examples of embodiments of the invention with reference to the drawing, which shows details essential to the invention, and from the claims. The individual details may be realised in an embodiment of the invention either separately or jointly in any combination.

Figure 1, a representation of multipath geometry,
Figure 2, a representation of guidance beam offset positions and multipath geometry,
Figure 3, some examples of guided object position accuracies for the measured position m_k,
Figure 4, some examples of guided object position accuracies for the processed position P_k according to the invention,
Figure 5, a block diagram of a realisation mode example of the multipath effect reduction device using the method according to the invention.

Figure 1 shows the multipath geometry and its effect on the beam.
An antenna A transmits a beam whose pattern is shown by Figure 1. This beam follows different paths: direct path R₁, reflected path R₂ (also called specular path). Due to differences in the path lengths of the direct R₁ and reflected R₂ waves, the beam following the direct path and the beam following the specular path will arrive with a (very small) time difference at the object O. This time difference translates into a phase difference. If the phase difference is small, then the two will add up to a larger value than from the direct path only, a phase difference near 180deg can lead to almost extinction. Whereas the range difference will vary with target distance, this interference or multipath effect will also show a range dependent fluctuation. The smaller the wavelength (higher frequency), the more peaks and lows will occur within a particular range bracket.
To enable the object O to determine its own position with respect to the main antenna direction, a number of predetermined beam position offsets can be introduced, either simultaneously as in a monopulse system or sequentially. Modulated by the beam pattern, the different beams will give different responses that the object uses to estimate its own position. For example in Figure 2, the direct paths R₁ for the dashed (offset beam C) and dotted (offset beam D) beams have the same gain and consequently receive the same amount of energy. As a result the object O is estimated to be at the antenna boresight.
The multipath effect, however, influences the accuracy of the estimation process. In particular because also the indirect paths I will have different angles (and gains) with respect to the bore sight of each beam position. Thereby, the multipath effect introduces additional errors in the position measurement. As an example, compare the dashed and dotted indirect arrows in Figure 2 that are shown different in size to represent different gains from the relevant dashed and dotted beam patterns. Although the object is at the antenna bore sight, now the indirect path contribution will induce an estimation error.
The objective is to separate "legal" (due to target and/or guided object manoeuvre) and "illegal" (due to multipath and/or measurement noise) measured position errors as much as possible. Taking into account the target manoeuvre capabilities as well as the guided object manoeuvre capabilities, the invention consists of limiting the influence of variations in the measured position error. The basic idea behind this limitation is that large changes in the measured position error can only be attributed to multipath and/or measurement noise, not to legal manoeuvres.
This limiting function can be range dependent, considering both target and guided object range. In this respect, the parameters of the limiting function are depending upon the maximum expected target manoeuvrability (dominant at longer range) and the guided object manoeuvre capabilities (dominant at short range). Different functions are applied for the vertical and horizontal direction in the guided object reference frame (note that due to rotation of this reference frame with respect to the local vertical, multipath effects can also be present in the horizontal direction).
The result achieved is a significant reduction in the systematic error (mean) as well as the random error (standard deviation), virtually without affecting legal guided object and guidance beam motions.
Figures 3 and 4 show examples of typical elevation accuracies as a function of guided object range for different bandwidths. Figure 3 shows these accuracies in terms of measured position m_k, whereas Figure 4 shows these accuracies in terms of processed position P_k. By processed position is meant the position obtained after the limitation according to the invention.
The X-axis represents the guided object range in metres and the Y-axis the elevation accuracy in 10^-3 radians. Actually, the guided object beam measured positions are usually angular data expressed in radians. Note that as in any radar system these data are not actually measured as such, but calculated in the guided object processor from measured voltage levels. An Analog-to-Digital Converter (ADC) provides the relevant digital equivalent to the processor.
The plain lines c₁ represent the evolution of the accuracies in terms of the mean value respectively for the measured position and the processed position with a bandwidth of 10%. The dotted lines c₂ represent the evolution of the accuracies in terms of the mean value respectively for the measured position and the processed position with a bandwidth of 3%. The dashed lines c₃ represent the evolution of the accuracies in terms of the standard deviation respectively for the measured position and the processed position with a bandwidth of 10%. The dot-dashed lines c₄ represent the evolution of the accuracies in terms of the standard deviation respectively for the measured position and the processed position with a bandwidth of 3%.
These two figures show that the processed position is more accurate whatever the bandwidth used is.
The method for the reduction of the multipath effect on the position measurement of a beam guided object according to the invention limits the influence of the variation of the measured position m_k into a predetermined interval [F₂, F₁]. As mentioned before, the predetermined interval [F₂, F₁] may be depending upon the maximum expected target manoeuvrability and the guided object manoeuvre capabilities which are range dependent, considering both target (Rtarg) and guided object range (R_amm).
In one possible realisation of this method for multipath effect reduction, the limitation comprises an estimation of the legal error measurement Δm_lim from at least the current measured position m_k and the previous processed position P_k-1. Then, the position P_k is processed by adding the estimated legal error Δm_lim to a reference position P_ref: P_k = P_ref + Δm_lim.
Figure 5 shows a possible implementation of the multipath effect reduction method according to the invention as variation limitation means (100). This variation limitation means (100) comprises a legal error estimator (110), which receives at least the current measured position m_k and the previous processed position P_k-1 and provides Δm_lim. This variation limitation means (100) comprises also a position processor (120) which adds the estimated legal error Δm_lim to a reference position P_ref: P_k = P_ref + Δm_lim.
The legal error measurement estimation Δm_k can be realised by, firstly, processing a variation by a subtraction between the current measured position m_k and the previous processed position P_k-1: Δm_k = m_k - P_k-1 Secondly, the variation Δm_k is limited within the predetermined interval [F₂, F₁] providing the estimated legal error Δm_lim as equal to:

Δm_k, if Δm_k is within the predetermined interval [F₂, F₁]
F1, if Δm_k is greater than F₁
F₂, if Δm_k is lower than F₂.

This first step can be implemented in a variation processor (111), and the second step in a processed variation limitation means (112). In this embodiment of the invention, the reference position P_ref is equal to the legal error estimation P_k-1 (initially at k=0, P_k-1 is set equal to m_k).
The delay T indicates that the difference Δm_k at instant k is calculated from the measurement sample m_k and processed position P_k-1 at instants k and k-1, i.e. separated in time by a delay determined by the measurement sample rate.
The two thresholds F_x (x=1,2) of the predetermined interval can be implemented by the following limiting function, which has been evaluated to work well (see Figure 3) : $F_{x} = maximum (\frac{k_{x_{1}}}{R_{t arg}}, \frac{k_{x_{2}}}{R_{amm}}) with x = [1, 2] and k_{x_{1}} > k_{x_{2}}$
The coefficients kx₁ and kx₂ of this function have to be tuned to the expected behaviour of the threat (e.g. highly manoeuvrable Anti Ship Missiles) and the behaviour of the guided object. Moreover, these coefficients obviously depend also upon the measurement sample rate.
As can be seen from the equation for F_x above, the limitation in the initial part of the guided object fly-out is determined by kx₂/R_amm, because at that time R_amm << R_targ. At that stage of the fly-out, the guided object will be gathering the guidance beam, target manoeuvres are less important due to the longer range R_targ: a target displacement over M meters will only result in a guidance beam displacement at the guided object position over M x R_amm/ R_targ metres.
At longer range when the guided object has been settled on the guidance beam, the target manoeuvrability becomes more important. The transition point is at kx₁/R_targ = kx₂/R_amm. Note that this has a meaning only when kx₁ >kx₂.
So initially the function value will decrease with time, but beyond the transition point, it will increase again until intercept.
The function described here is not necessary the only possible implementation. Other functions may work equally well or even better than the one described here. The basic idea of the invention is that "legal" error measurements can only be caused by target or guided object manoeuvres and that such manoeuvres will be less than a certain maximum. Obviously this maximum depends on the capabilities of both the target threat and the dynamics / kinematics of the guided object, both of which are range dependent.
The introduction of range dependent limitations on the increase and decrease of the beam rider position measurements reduces the influence of multipath errors to an acceptable low level. By this way, the use of a conventional type of transmitter with bandwidth of less than 3% is allowed and the need for additional filtering which will affect the guided object stability is obviated. Therefore, the invention is a low cost solution providing a good guided object stability.
The multipath effect reduction method can also be implemented as additional software in the existing board computer processor of the RF guided object control system.
More generally, such multipath effect reduction method may be used by any kind of beam riding object control system, as for example guided ammunition and missiles.

Claims

Method for reducing the influence of the multipath effect on the control of a beam guided object, comprising:
- the beam guided object measuring its own position with respect to the guidance beam, said measuring being possibly affected by multipath effect;

- determining a variation of said measured position with regard to a previous processed position;
the method being characterized in that it comprises:

- limiting said determined variation into a predetermined interval in order to limit the influence of multipath effect;

- controlling said beam guided object based on said limited variation.