GB2062395A - DME System - Google Patents

Abstract

A DME system includes an interrogator (Fig 17) aboard an aircraft transmitting an interrogation pulse pair having predetermined waveform characteristics and a ground- positioned transponder (Fig 18) receiving and retransmitting said pulse pair after a fixed time interval, with interrogator receiving means (31-39) responsive to retransmitted pulses from said transponder, comprising circuits (38) for evaluating range as a function of the time delay between interrogation and reception at said interrogator of said retransmitted pulses; a filter (45) within the interrogator and/or transponder receiving means for increasing the slope of at least one of the pulses of said pulse pair passing therethrough; and a pulse shaper (25) within said interrogator and transponder for modifying the waveform of the pulses of said pulse pair prior to transmission to reduce their radio frequency spectrum bandwidth. The transmitted and received pulses have exponential rising and falling edges and are converted by the filters in the receiving means to rectangular envelope pulses. Distortion of received pulses due to ground echos is removed. <IMAGE>

Description

SPECIFICATION DME System The present invention concerns electronic distance measurement or (DME) systems for air navigation.

As known, DME is a distance measurement system used in air navigation for relatively short distances, in which transmit/receive equipments positioned on the ground and on board the aircraft are used.

Pulses of a predetermined frequency are transmitted from the aircraft interrogator and received by the ground transponder station which replies to them on another frequency after a predetermined time delay. The time delay between the transmitted pulses and those received by the interrogator on board is the parameter employed for measuring the aircraft's distance from the ground station.

Accordingly, DME, which is based on the secondary RADAR principle, enables measurement on board the aircraft of the distance between an aircraft and the ground transponder ("range").

The signals transmitted by the interrogator and the transponder consist of a carrier, amplitude modulated by a pair of pulses.

The distance is determined by the time-lag between the transmission of an interrogation pulse pair from the aircraft and the reception of a corresponding pair of reply pulses transmitted from the ground station. The time spacing between the pulses of the pair also characterizes the mode of operation (operation X or operation Y with a time spacing between the pulses of 1 2 or 30/36 microseconds respectively). These spaced pulses are recognized by means of a decoding circuit.

As known, the transmitted pulses are basically Gaussian, since the Gaussian wave-form occupies a minimum spectrum for a given wave (pulse) energy. That is, a Gaussian transmission has the advantage of having a compact spectrum. Nevertheless, it has the disadvantage that the arrival instant of the pulse is not easily recognized, as would be the case with a rectangular wave.

In a DME system it must be possible to recognize the arrival instant with precision, as an error of only one microsecond would result in a distance error of about 1 50 meters.

The recognition method for pulse arrival instant laid down by the International Civil Aviation Organization (ICAO) is to "intersect" the Gaussian pulse at half amplitude, i.e. when the peak value of the Gaussian pulse has been established, the latter is intersected at half amplitude, measuring the arrival instant corresponding to this point. This, however, presents considerable difficulty because the pulse may arrive either very weak or extremely strong depending on aircraft-to-ground station range.

The amplitude variation may be as much as 80 dB, (a variation of 104). Due to this extremely wide range of signal strength values, a linear receiver obviously cannot be used for Gaussian pulse reception, since the receiver must not distort pulses at all. A linear type receiver with such a dynamic range would not be technically feasible.

In DME systems of prior art, the above-mentioned problem has been solved through addition of a logarithmic receiver and/or amplifier, which supplies an output voltage proportional to the logarithm of the input voltage. Thus two receivers and/or amplifiers are employed, i.e. a linear receiver with controlled variable gain and a logarithmic receiver. The logarithmic receiver controls the gain of the linear receiver in such a way that its output signal level is always constant. This control is a conventional one and is also common in radio receivers generally.

As we have seen, in known DME receivers, the constant level supplied by the linear amplifier under the control of the logarithmic amplifier, corresponding, for example, to 1 volt, is "cut" in half. A comparator therefore samples the signal at 0.5V to obtain the arrival instant of this signal. Naturally, to be able to give the logarithmic amplifier/receiver the time to control and attenuate the linear amplifier/receiver in relation to the arrival signal amplitude detected, a delay line (with a bandwidth on the order of 10 MHz) line must be placed before the linear amplifier, i.e. for correct reception, the logarithmic amplifier must be capable of sensing the peak value of the input signal. When this has been detected, the logarithmic amplifier controls the linear amplifier gain.

It is evident that the two receiver and delay line circuitry is relatively complex.

From the foregoing, a serious disadvantage will be seen relating to known DME receivers range errors arising in the function of detecting or "reading" the input signal.

It is well known that distance precision could be improved if pulses with short rise times were used. These steep pulses would improve distance precision in two ways, i.e. the "decision" level would be reached sooner and the crossing of the "decision" area would be steeper. Consequently, the leading edge would be less affected by nwulti-path effects and any variation in the "decision" level would produce a smaller range error. While pulse rise time is significant, pulse delay time has no influence on distance precision and may be traded off to optimize other characteristics.

Other drawbacks of known DME receivers concem the other two main functions of these receivers, i.e. adjacent channel rejection and echo suppression. These functions will be briefly described below as this invention, as will be seen in more detail later, proposes the implementation of improvements applying to all three of these functions, the first of which, as seen, is the reading of input signals.

A DME receiver is preset to receive a certain frequency, e.g. 1125 MHz, and must be able to reject any interrogation which arrives on adjacent channels, e.g. 1 24 or 1126 MHz, irrespective of the amplitude of such an adjacent channel signal. This translates to a requirement that the receiver preset at 1125 MHz must be capable of receiving signals on the order of -90 dBm, while rejecting signals of O dBm on adjacent channels. This imposes a high standard of selectivity performance on a DME receiver. In prior art DME receivers, adjacent channel rejection depends on extremely selective complex filters which are difficult to design and implement and also tend to add complexity.Finally, as regards multi-path (echo) suppression, previously mentioned, this problem is briefly described as follows. A DME receiver positioned on board an aircraft approaching a ground station receives not only the direct signal from the ground station, but also echo signals from ground or other fixed features. These are generally less intense than the direct signal, however the DME receiver must be capable of rejecting these multi-path signals. In known DME receivers, this function is also fulfilled by means of fairly complicated circuitry, which contributes to the complexity of the whole receiver.

To improve the first above-mentioned function, i.e., detection of the arrival instant of signals in DME systems, many proposals have been put forward, in particular that of modifying the pulses transmitted, i.e. making their leading edges steeper. The use of steep (fast leading edge rise) pulses in the transmission phase is presently unacceptable, as present day systems foresee 252 adjacent channels, with 1 MHz channelization. The inherently wider spectrum produced by such fast rise times would cause interference to the adjacent channels. Since DME is a widely used standard system, any alteration of standard signal formats or the introduction of new signals would be required to be compatible with existing installations.

To be more precise, all proposals regarding transmitted pulse modification must consider the problem of adapting the time shape of the pulses to reach the desired signal-noise ratio without violating the present DME spectral limit specifications.

In some proposals, at least the first pulse in the pulse pair would not meet the time tolerances presently foreseen for the DME. This may create icompatibility problems. An existing interrogator must, of course, be capable of using a newly conceived transponder, measuring range with greater, or at least the same, precision as at present, and, in the meantime, a new type of interrogator must be capable of using an existing transponder measuring range with greater, or at least the same, precision as at present.

The above incompatibility is encountered both in cases where the two pulses are modified and in cases where only the first is varied. In the latter case, in fact, it is also necessary to take into account time spacing tolerances of the two pulses, which influence a correct signal recognition in the channel.

In the final analysis, it may be observed that, if the temporal wave-forms over-deviate from the present ones, this will seriously compromise operation with the existing DME equipment. Therefore, a solution for DME improvement which does not involve drastic steepening of the leading edge of the transmitted pulses would appear particularly attractive, both for the abovementioned compatibility reasons and because it would permit use of high powers during transmission, the spectral purity restrictions being not critical.

In the light of the above-mentioned limitations in known DME systems the purpose of this invention is to provide a type of DME system capable of optimizing the three above-mentioned functions, namely the reading of input signals, the rejection of adjacent channels and multi-path suppression by means of an extremely simple, compact, reliable circuitry.

A DME according to the invention can be implemented with an extremely small number of compact integrated circuits, easily available on the open market, and can be made relatively insensitive to circuit component variations and tolerances.

According to the present invention there is provided a DME system including an interrogator located aboard an aircraft having transmitter means for transmitting an interrogation pulse pair having predetermined waveform characteristics and a ground-positioned transponder having means for receiving and retransmitting said pulse pair after a fixed time interval, comprising:: interrogator receiving means responsive to retransmitted pulses from said transponder, said interrogator receiving means comprising circuits for evaluating range as a function of the time delay between interrogation and reception at said interrogator of said retransmitted pulses; first means within at least one of said interrogator receiving means and said transponder receiving means for increasing the slope of at least one of the pulses of said pulse pair passing therethrough; and second means within said interrogator and transponder for modifying the waveform of the pulses of said pulse pair prior to transmission to reduce the radio frequency spectrum associated therewith, said second means being associated with each of said receiving means including said first means.

Further characteristics of the DME system according to the invention will appear clearer from the following detailed description of a preferred implementation of the same, with reference to the accompanying drawings, in which: Figure 1 is a block diagram illustrating the operational principle of a DME system, Figure 2 is a diagram of wave4orms explanatory of the operation of the DME system of Figure 1, Figure 3a is a block diagram illustrating a device for modifying the transmitted wave-form, as proposed in the prior art, Figure 3b is a block diagram illustrating a typical implementation of a device for modifying the received wave-form, according to the criteria of the invention, Figure 4 is a general overall block diagram of a DME system according to the present invention, Figures 5 and 6 illustrate characteristic wave-forms of a DME system according to the invention, Figure 7 illustrates a pulse-shaper for use in the DME system according to the invention, Figure 8 is a graphical illustration of the electrical characteristics of the pulseshaper of Figure 7, Figure 9a is a graphical illustration of the wave-form of the input pulse, entering the pulse-shaper of Figure 7, Figure 9b is a diagram illustrating the wave-form of the pulse as shaped by the pulse-shaper of Figure 7, Figure 10 is a graph illustrating the spectral characteristics pulse of the pulse of Figure 9b, Figure 11 is another graphic illustration of a pulse of smoothed shape, with better spectral profile than that of Figure 9b, obtained with the use of two pulse-shaping circuits of the type illustrated in Figure 7 in cascade connection, Figure 12 is a general circuit diagram for an anti-multipath filter in base band used in the DME system according to the invention, Figure 1 3 is a graph illustrating the characteristics of the anti-echo filter of Figure 12, Figure 1 4a is a graph illustrating the filtered pulse obtained by using the filter of Figure 12, Figure 1 4b is a further graphic illustration of both the input and output pulses to and from the filter of Figure 12, Figures 1 5a to 1 sic are wave-form diagrams illustrating the effect of multi-path reception on the characteristic pulses of the DME system according to the invention, Figure 1 6a is a table in which data are shown, relevant to the filter of Figure 1 6b obtained with both variable and constant gains, Figure 1 6b illustrates a basic diagram of a Kalman type filter used in a P-DME system according to the invention, Figure 1 7 is a block diagram of the interrogator in DME system according to the invention, and Figure 1 8 is a block diagram of the transponder in a DME system according to the invention.

Referring now to Figures 1 and 2, to make the invention more easily understandable, the general form of a DME system will first be described. As mentioned, the conventional DME system is mainly based on the principle of secondary radar even if the ground and air roles are inverted. Two fundamental equipments are included in Figure 1; an interrogator 1, aboard the aircraft, and a transponder 2, on the ground. The main components in the interrogator 1 are, antenna Al, a pulse generator 3, transmitter 4, receiver 5, decoder 6 and measurement circuitry 7. The ground transponder 2 is mainly composed of; antenna A2, receiver 8, decoder 9, delay line 10, modulator 1 transmitter 12 and pulse generator 13.Each aircraft transmits a signal to ground composed of (for example) an Lband carrier amplitude modulated by a pair of Gaussian pulses, which transponder 2 on the ground receives and retransmits after a predetermined delay. On board the aircraft, the signal is received and the range information is determined from the time-lag between transmission and reception. Since a transponder replies to may different aircraft simultaneously, each aircraft must be capable of recognizing its own replies. For this purpose, the interrogator's mode of operation is generally divided into two phases.

In one, the search phase of operation is based on the fact that, since each aircraft interrogates with a random repetition frequency around a nominal value, it is very improbable that two replies for another aircraft will be received with the same delay. This principle is illustrated in Figure 2. Therefore, when the aircraft is connected to any arbitrary transponder, it recognizes its own replies.

It must be observed that, in this search phase, the interrogation frequency is high (about 1 50 interrogations/second) and the aircraft will therefore have moved very little between interrogations.

Taking into account the information lost due to transponder efficiency, approximately 70%, most of the present interrogators consider the reply valid when three coincidences out of five successive interrogations occur, and the circuits are adapted to this "data good" rate.

The second interrogator operative phase is the "track" phase. Once the aircraft has made the first measurement, operation is simplified. Since there is little difference between two measurements, the interrogator always waits for the next reply within a delay very close to that previously measured; a "gate" is then opened around the previous delay and only the reply with said gate is received. A lower interrogation frequency is sufficient in this phase (30 interrogations/second). Figure 2 shows 5 consecutive tracks obtained on an oscilloscope, with triggers controlled by the aircraft interrogation. If one considers, for example, a maximum distance of 350 km, the reply will certainly be within 2400 microseconds. The transponder transmits 3,000 pulse pairs per second and in this time interval the interrogator will receive an average of 7 pairs, only one of which is the genuine reply.As each pulse lasts 3.5 microseconds (for example), the ratio between the time the signal is present and the time only the noise is present is 0.02. It is therefore seen that it is very improbable that two different interrogations (non-valid pulse pairs) will have the same delay (in Figure 2 this happens only once). In the third track, the absence of a valid reply may also be noted. This may occur if the transponder has replied to another aircraft shortly beforehand. After reception of each interrogator, the transponder has a typical dead time of 60 microseconds to keep its duty cycle from below a nominal value. The RF band allotted to the DME is 962 to 1213 MHz with channels spaced 1 MHz from each other. There is always a standard difference of +63 MHz between the reply and interrogation frequencies.

To obtain a high traffic capacity, the channels available are reached by multiplexing. These are divided into: (a) X channels: the spacing between the two pulses in the pulse pair is 12+0.25,us and the coupling between the interrogation and reply frequencies as follows: Interrogation (MHz) Reply tMHzJ 1025to 1087 925to 1024 1088 to 1150 1151 to 1213 (b) Y channels: the spacing between the two pulses in the couple is 36+0.25 us for interrogations and 30+0.25 pts for reply; the coupling between the interrogation and reply frequencies being as follows: Interrogation (MHz) Reply (MHz) 1025 to 1087 1088 to 1150 1088 to 1150 1025 to 1087 The present pulse has the following features::- rise time (10% to 90%)=2.5+0.5 gs=fall time duration (50%)=3.5+0.5 prs spectral purity defined by limiting the effective radiated power (ERP) during a pulse to four frequency windows 0.5 MHz wide centered +0.8 MHz and +2 MHz with respect to the carrier.

One of the main factors influencing measurement precision is the method in which the signal's arrival instant is detected, both at the transponder, which must introduce a fixed delay, and at the interrogator, which must measure the reply delay. The method used at present in ICAO regulations is, as seen, to refer to the instant at which the leading edge of the second pulse reaches 50% of the peak value. As already mentioned, it is always necessary to take into account the effect of multi-path reception on the useful pulse, i.e. wave-form detection, which provokes a variation in the link between the recognition instant of the pulse arrival and the estimate of the propagation time as a function of range. This effect depends upon the detection system selected; in particular, with the above-mentioned detection method at 50% of the peak value, the multi-path effect is two-fold.On one side, it can vary the peak value in relation to the remaining wave-form, and, on the other, operating directly in the pulse area near the detection threshold, it can alter the crossing point of said threshold. Indeed, one significant purpose of the invention is to reduce the error caused by multiple paths, with modifications mainly in reception without altering the present basic DME philosophy or transmission format. Among these modifications the following are fundamentally important and are consistent with safeguarding compatibility with the equipments already operating: (a) The use of a filter in reception capable of making the leading edge of the input pulse steeper, and for substantially eliminating the impact of multi-path received signals or the said leading edge.

(b) The use of the first pulse in the pulse pair to estimate the arrival time, as this is the one least affected by multi-path signals (the second pulse of the pair is subject to multi-path echoes provoked by the first as well as its own transmission).

(c) The adoption of a detection threshold value, reached by the pulse in the shortest possible time, compatibly with the signal-noise ratio concerned, as, in this way, operation takes place in a less echo-distorted area.

(d) The use of detection methods which do not rely on the maximum wave-form value as a reference to position the threshold, since that maximum value is quite affected by echoes.

(e) Causing the receiver to operate at a wide band when it is confirmed that it detects a pulse on the correct channel, this avoiding further local distortions, but creating the necessity to use a cooperating circuit capable of correlating the received pulse pair against that of the corresponding interrogator (selected channel) (Ferris discriminator or similar).

(f) To carry out an appropriate numerical filtering after detection for reducing highly variable errors between successive measurements (Kalman filter or derivatives).

Having taken into account the general configuration of an existing DME equipment, and always for purposes of compatibility with it, to obtained the aforementioned objects, the invention is based, as regards the over-all errors due to the various causes mentioned, on the basic specification suggested by the FAA, with an over-all error of 100 feet (30 meters) on the signal's two-way-path, including a multi-path echo signal 6 dB lower than the main one, with each carrier phase and each delay (referred to the worst case). This specification is intended to correspond to the critical position, after which it is no longer permitted to interrupt the landing phase as defined at 50 feet (1 5 m) from ground height and 15,000 feet (4,5000 m) distance from the DME antenna.

These FAA indications have been taken as a starting point to define the precision specifications for the P-DME in this invention.

The terms eR, and eRT are therefore defined as noise errors for interrogator and transponder respectively; similarly ep and epT are the polarization errors. The error eE2 due to echo on the two-waypath will therefore be:

where all values are expressed in meters.

Following the prescriptions of the FAA, the specification for measurement accuracy are taken to be: eR=eRT=30 feet (9 m) ept=epT=35 feet (10.6 m) eE2 eye1= =38 feet (11.6 m) 2 where eE, indicates the error due to multi-path echo signals on the one-way-path.

It is noted however, that further practical and constructive considerations could suggest a different division of values among eRI, eRT, ep, and epT.

With specific reference to figures 3a and 3b, a description will now be given of the inventive principle employs to reduce errors due to multipath, based on the observation that the DME signal between transmitter output and demodulator input is subject to linear operations only. This means that the wave-form modification required to reach the required'precision may be carried out in the receiver itself, a novel concept vis-a-vis the prior art solution which aimed at modifying the wave-form of the pulses transmitted. Fig. 3a shows the baseband equivalents of the linear part of the prior art system, i.e.

with modification in the transmitted wave-form.

In principle, this modification may be considered as having been obtained through a linear network with the following transfer function: x'(f) A'(f)= x(f) It appears evident that, for the purposes of the received signal, y(t), it is equivalent to transfer the A' network to the receiver, according to the invention (Figure 3b) with the sole restriction that the following receiving filter must be feasible. R(f)=A'(f) Q(f). This resolves the above-mentioned compatibility problem, without modification to the transmitted wave-form x(t). The system and equipment of the invention also presents further advantages as regards signal treatment. Firstly, it permits more radical wave-form modifications, since these do not affect the transmitted pulses.

Assuming 1 A'(f)= A(f) then the input signal of the Q(f) network is the start pulse g(t) which has a very short rise time and therefore practically no errors; therefore Q(f) needs only to introduce a small pulse modification to have a small error.

It is also evident that x(t) presents a much narrower spectrum than x'(t). Therefore, as both waveforms must respect the ICAO spectral purity standards, it is clear that the present system is capable of radiating a higher power than systems with transmitted pulse modification. The system or equipment could therefore, if necessary, be used in two ways: on short range as a P-DME and on longer range as an ordinary DME. As regards the system's behaviour when faced with noise errors, it will first be noted that, as it must make the edges of pulse x(t) steeper, the A' (f) network has a high-pass type characteristic which intensifies the high frequency components in the spectrum. It is therefore clear that, y(t) being equal, the noise power at the receiving filter output is higher in the invention system.

Naturally, it must be borne in mind that, as seen, due to spectral purity restrictions, it is possible to have V > V'. Therefore, the power radiated in the spectral windows prescribed by the ICAO being equal, the present system may also provide a better signal/noise ratio at the demodulator input than that in a system in which the transmitted wave-form is modified according to the prior art technique. Therefore, the basic concept of this invention is, as will appear evident, to modify the wave-form in the receiver instead of that in the transmitter.

Some forms of DME system implementation will now be described which enable the required performances to be reached using fairly simple implemented linear networks.

in this connection, refer to Figure 4, showing an implementation of the system or equipment (one-way only), according to the teaching of the present invention.

A pulse generator 14 produces pulses which are assimilable to ideal square waves. This is followed by a pulse shaper 15, whose output x(t) meets the ICAO specifications. After amplitude transmitter-modulator 16, the signal meets the ICAO spectral purity rules, and this restricts the maximum power which can be radiated (maximum ERP).

In reception, after the amplifier block and frequency conversion 17, we have the filter 18, its function being to reduce the error due to multi-path echoes. At the output of filter 1 8, demodulator 1 9 is presented a sharply rectangular envelope. P(f) will indicate the equivalent low-pass filter 18.

Decisional circuitry 20 is cascade-connected to demodulator 19.

In the search for the functions A(f) and P(f) a general optimization method has not been followed, but it has been assumed that feasible structures are provided, as will be further described hereinafter.

Demodulator 19 is a conventional envelope detector.

A 10% threshold of the maximum pulse amplitude has been selected, however as will be described in more detail later, this is subject to modification. For the shaper A(f), the following transfer function applies:

The following values have been chosen for the two parameters B1 and B2: B1=O.1 15MHz B2=1.4MHz The filter P(f) is composed of the cascade of a high-pass network A'(f) and low-pass network Q(f): P(f)=A'(f) Q(f) For Q(f) a 2-pole Butterworth filter with 3 dB cut-off frequency B3=1 .5 MHz has been selected.

The most suitable A'(f) network, considering the chosen A(f) shaper, has the following transfer function:

where: S4=0.1 15 MHz B,=l.l MHz The received wave-form has the y(t) envelope shown in Figure 6 to which a rise time (10% to 90%) of t=0.4 5 ys applies.

In these conditions, with the chosen echo model, the maximum error provoked by multipaths is: eye=30 feet (9 m) consistent with the specifications.

The thermal noise performances are evaluated on two factors:- the threshold-to-noise ratio must guarantee a prefixed false alarm rate; the noise error must be contained in the limits prescribed by specifications.

As regards the first factor, a false alarm rate of one second may be assumed. In the system of the invention, this imposes the following threshold-to-noise ratio: T -12 dB N As regards the error due to noise, this is calculated according to the classic formula and the following limitation is imposed: eR < 30 feet (9 m).

Of the two limitations in the case in question, the per limitation is the more stringent, and imposes the following threshold-noise ratio: T -=12.2dB N with a corresponding signai-noise ratio of: S -=32.2 dB N At this point, having evaluated on the basis of the receiving filters, the noise power which would be obtained in base band for a noiseless receiving apparatus, the overall margin of M dB for the proposed system is obtained, considering the attenuation of free space at the distance of 1 5,000 feet (4,500 m) from the DME antenna, as per specification. It has also been assumed that the receiving antenna has 0 dB gain and a noise temperature equal to the reference environmental temperature.The value obtained is: M=21 dB This margin may be divided into two parts, the first M, will take into account the deterioration of real propagation with respect to the ideal one (fading, radiation pattern, etc.), the second, M2 will take into account the deterioration of the signal-noise ratio in the real receiving chain (noise figure).

This value makes this system's operation seem satisfactory, even in is first implementation form.

It should, at this point, be remembered that numerical filtering is employed in the interrogator, according to the invention which will be described more fully later, and it may therefore be presumed that its behaviour with respect to error due to noise is even better than that above-mentioned.

As previously indicated, the threshold values established on the basis of the two distinct considerations, are not far apart. On the other hand, the system thus configurated is already within the specifications before each data processing. The use of the Kalman filter, according to the invention, should therefore be considered a feature of the apparatus, with the sole restriction that there must never be a greater error after filtering than the system error.

Tests carried out varying filter parameter values have confirmed that this solution is not critical and transfer function variations, in a rather wide range, have little effect on the values of eR and eE.

A more accurate tolerance evaluation may be made after selection of a definitive filter configuration.

To obtain experimental confirmation of the validity of the operational principles previously illustrated, a test program has been conducted, some selected results of which are described hereinbelow.

Experiments have been made on the circuits implemented in this phase, which have confirmed the previous numerical evaluation made by a computer.

A first experimentation phase was carried out in base band to verify the practical reliability of the proposed signal handling.

Referring to Figure 7, a shaping circuit of the T-bridge type is shown. That network, consisting of two resistances R, inductor L1 and capacitor C1 is of low-pass type, with a 3 dB, 100 KHz band (on the basis of the chosen values and referring to the characteristic of the shaper in Figure 7, illustrated in Figure 8) matched in input and output for an impedance of 50 ohms.

With the input signal shown in Figure 9a, the described circuit supplies an output wave-form shown in Figure 9b. The output signal spectrum, illustrated in Figure 1 is relatively favorable compared to spectral specifications. It may be further improved by inserting another circuit of the type shown in Figure 7 in cascade, but with different parameter values. The result may be seen in Figure 11, which shows a pulse similar to the previous one, but with smoothed shape and consequently better spectral characteristics.

The multi-path anti-echo filter used in base band is a 4-pole Butterworth low-pass filter with 2.17 MHz bandwidth, cascade connected to a band-pass T-bridge filter with center frequency fro=1.25 .25MHz and attenuation outside the band of 16 dB. The dual circuit unit is illustrated in Figure 12. To be more precise, and with reference to Figure 12, the low-pass section includes the inductors L1 and L2 connected in series and the capacitors C1 and C2 connected as shown between the L1 and L2 inductor terminals respectively and ground.The band-pass section, cascade connected to the low-pass section, instead includes four resistances R, R, R1 and R2, connected to the point connecting Rand R, the other terminal of R2 being connected to the parallel connection of a capacitor Ca and inductor L4. Connected in parallel to the R resistances are a further resistance R, and a capacitor C3 and inductor L3 in series.

The parameters for implementations of this filter are as follows: L,=2.8 ,H L2=6.78 ,uH L3=3.40 ,uH L4=1 1.3 ,uH C,=2.70 nF C2=1.10 nF C3=4.70 nF C4=1 .40 nF R=50 ohms R=7.45 ohms The result of this filter implementation is the frequency characteristic shown in Figure 13. The shaped pulse, which is passed through this filter, provides the output pulse shown in Figure 1 4a, and has a rise time tis=0.2 ,us and a duration td=3.4 ,us, thus obtaining an improvement in rise time by a factor of 10 (see Figure 1 4b), thus confirming this solution's validity.

Tests have been made to verify, at least quality-wise, the benefits deriving from the introduction of the receiving filter on a pulse affected by multi-path echo. The results are shown in Figures 15a to 1 sic. More particularly, Figure 1 spa shows the filtered and unfiltered pulses without echo: Figure 1 sub shows the pulses affected by an echo with equal phase, amplitude 0.5 and delay 1.5 microseconds with respect to the direct signal: and Figure 1 sic shows pulses affected by echo with opposite phase, amplitude 0.5 and delay 0.5 microseconds with respect to the direct signal. As will be noted, with equal echo delay and amplitude, filter introduction has notably preserved the leading edge useful for detection.

A description will now be given of the system and numerical filtering means according to the present invention criteria. This numerical filtering of range measurements (aircraft-transponder distance), which must be carried out by the interrogator on board, constitutes a new feature fo the invention as compared with conventional air navigation apparatus, a peculiar characteristic of the proposed P-DMF system. The purpose of numerical filtering described hereinbelow, is to reduce the error due to noise affecting the measurements in the P-DME system.

Due to the need to work in real time, the type of numerical filtering proposed must be carried out in the interval At (approximately 33 milliseconds) between two successive measurements. This can be viewed as a constraint on the algorithm complexity of the filter. Therefore, the choice made for the P DME was to use the Kalman filter category which, compared to other known solutions in the radio assistance field, afford the desired simplicity.

This type of filter is capable of providing an estimate of the range and range-rate of the aircraft on the basis of the present measurement and of the projection of the last estimate made. As may be shown, this, is in practice, equivalent to taking all previous measurements into consideration, without, however, having to resort to an increasing memory loading.

To project the last estimate, the Kalman filter requires a mathematical model of the aircraft dynamics, which, in the simplest instance, takes into account only the range and range-rate (model with bidimensional state vector), and, if necessary, may also contain a certain number of derivatives of a higher order. A model with bidimensional state vector is already capable of providing a considerable reduction in error due to noise, despite the fact that further improvements appear certain on the basis of a more accurate aircraft description.

In the case of the P-DME, this type of filtering is, however, unable to act upon errors with a very high correlation time with respect to the interval At and cannot therefore be considered a solution either for errors caused by polarization or by propagation. As regards the latter, it must be borne in mind that, although it is generally necessary to proceed differently, in cases where the errors caused by multi-path have not too high a correlation time, the Kalman filter will even further reduce the error connected with it.

An evaluation has, in the meantime, been made of a Kalman filter with bidimensional state vector, the block diagram of which is illustrated in Figure 1 6 and for which the following expression has been found relating to the reduction factor of the range error in steady connection:

in which eR=2 vmeasurement error before numerical filtering e1 R =2Rmeasurement error after numerical filtering.

At=33 ms, generally the interval between two successive measurements and aM the maximum aircraft acceleration (intended as range-rate derivative).

An example is given by the table in Figure 1 6a which shows the error reduction obtainable at corresponding values of aircraft measurement error eR and maximum acceleration aM, easily verified in the case in question. In the above-mentioned table g=32.2 ft/s2 (9.81 m/s2) which is the gravity acceleration.

These results may also be applied to a stationary filter, i.e. the filter for which the gains Ak and pk in Figure 1 6 are constant and equal to the steady values in the non-stationary version.

In this case, the advantage of a lower calculation time required to carry out filtering is paid in terms of response time to manoeuver type stress, which is normally above that obtainable with a variable gain filter.

From the above it will be clear that it is convenient to use the numerical filter described to process measurements before presentation. A calculation network is necessary for this filtering in the present state of availability of digital components. Microprocessors, make it possible to provide this calculation network simply, economically and reliably. On the other hand, as will be shown herebelow, the use of the microprocessor in the entire system's control phase offers a brilliant solution to system and precise problems. The choice to introduce these components in the implementation of both the equipment on board and that on ground therefore appeared decidedly justified and has a determining influence on the architecture of the equipments.

A detailed illustration will be given below of the structures of the ground and on-board equipments illustrated in the general block diagrams in Figures 1 7 and 18, respectively.

Figure 1 7 shows the block diagram of the equipment on board. Note how the various sections are handled by the microprocessor 22, which controls the various operations in correct sequence to measure distances and make suitable calibrations, and to give the pilot a diagnosis of the overall equipment functionality, in addition to the numerical filtering already mentioned. A high stability clock generator 23 provides timing to both the microprocessor and the circuits generating pulses or measuring delays. The presence of these circuits, even if apparently contrasting with the presence of a processor on board, is a consequence of the speeds at which they must operate.A microprocessor with instruction times typically in the microsecond range would not, in fact, be capable of recognizing the exact arrival instant of the signal and it is therefore indispensable to use faster circuits for these functions. After having received a trigger pulse from the microprocessor 22, the couple generator 24 generates both pulses in the couple of interrogations: the pair of pulses thus obtained is sent to a shaper 25 which shapes each single pulse, as previously described. The shaped pulses then pass to a modulation amplifier 26, a final power stage 27 and a circulator 28. The carrier frequency is generated in a quartz-controlled synthesizer 29 which may be programmed by means of digital switches on the control panel 30 and operated by the pilot, who can thus choose the transponder on which to measure distance.The same signal, with suitable levels, is also sent to the reception mixer. Since there is a difference of 63 MHz between the transmission and receiving frequency in the DME system, this is the value of the intermediate frequency. At the mixer 31 output we therefore have 63 MHz signals, the envelope of which represents the pulses emitted by the transponder.

Figure 1 7 also shows three ampiification blocks; the preamplifier 32, the intermediate frequency amplifier 33 and the intermediate frequency amplifier 34. The purpose of the first two amplification blocks is to improve the equipment's noise figure; the criterion of economic optimization will determine if one of the two should be suppressed. As regards the main IF amplifier 34, one solution is the use of a gain-controlled amplifier in which the signal is subject to a compression which is a function of the instantaneous input value. With suitable design, a variation of many dB in the input signal may be translated at the output into a variation of a few dB.

The final choice will, however, be closely connected with the type of detection selected.

Figure 1 7 shows the channel detector 35, which, as previously mentioned, is necessary since the receiver operates at wide band. The signal leaving the comparator 36 is sent to the window circuit 37, which, like other inputs, receives the output from the channel detector 35 and the time enabling provided by the microprocessor 22 and the counter 38. The decoder 39 then recognizes if the input pulse pair has the correct time spacing. If so, the signal leaving the decoder 39 stops the counter 38 and signals to the microprocessor 22 that the measurements cycle is terminated.

With reference to the block diagram in Figure 18, the electronic ground equipment, or transponder, is, in principle, fairly similar to that on board. There is, however, a slight difference since different aircrafts must be simultaneously served. Above all, for reliability, the entire assembly is duplicated, and, when out of order, the automatic switching on the equipment is indispensable. There are also accessory circuits, e.g. the identification pulse generator, the dead time circuit 41, which inhibits transponder operation for a fixed time (approximately 50 microseconds after an interrogation has been accepted as valid), and the main delay circuit 42, which maintains the overall delay time of the transponder rigorously fixed (50 microseconds).

Therefore, without dwelling on the parts already described for the interrogator on board, it is opportune to note how the intermediate frequency IF stage, composed of the mixer 43, the first intermediate frequency amplifier 44, the filter 45 and the second intermediate frequency amplifier 46 operate on signals which are very different among them and how imperative it is to use amplifiers which are capable of instantly adapting themselves to working conditions. In choosing same, considerations similar to those made for the equipment on board are valid. Having taken into account the relatively limited number of transponders compared to that of interrogators, it may be seen that it is best to optimize the ground equipment as much as possible, even by applying techniques which would be un-economical if used on the equipment on board.

It is also convenient in the transponder to use numerical control circuits providing compensation delay times in very small steps, and to monitor the various transponder parameters (RF power, delay times variations, etc.), provide the identification signals of the transponder itself and register any faults noted on special physical support, if necessary sending suitable alarm messages to remote operators.

The employment of the synthesizer 47 also in the transponder, as this operates on one frequency only, proves convenient in view of the alignment difficulties encountered in conventional frequency multiplication circuits. Furthermore, a relatively high foreseeable production of synthesizers for use in air navigation should make the latter competitive, from a merely economic point of view.

It will be noted at this point that an air navigation DME system according to the invention is provided in which, owing to the described wave-form modification, the three functions (reading the arrival signal, and, more specifically, its instant of arrival, rejection of adjacent channels, and echo suppression) are optimal and preserve the structure of the standard DME system. Moreover, the number of DME channels is not changed. Unlike in known structures in which this modification is carried out in transmission, here it is made in reception, i.e. it is not necessary to drastically steepen the leading edge of the transmitted pulse.

Claims (11)

Claims

1. A DME system including an interrogator located aboard an aircraft having transmitter means for transmitting an interrogation pulse pair having predetermined waveform characteristics and a ground-positioned transponder having means for receiving and retransmitting said pulse pair after a fixed time interval, comprising:: interrogator receiving means resonsive to retransmitted pulses from said transponder, said interrogator receiving means comprising circuits for evaluating range as a function of the time delay between interrogation and reception at said interrogator of said retransmitted pulses; first means within at least one of said interrogator receiving means and said transponder receiving means for increasing the slope of at least one of the pulses of said pulse pair passing therethrough; and second means within said interrogator and transponder for modifying the waveform of the pulses of said pulse pair prior to transmission to reduce the radio frequency spectrum associated therewith, said second means being associated with each of said receiving means including said first means.

2. A DME system according to claim 1 in which said first means are provided within the receiver circuits of both said interrogator and said transponder, said first means being operative to present said circuits for evaluating range with pulses having decreased rise time prior to said second means.

3. A DME system according to claim 1 in which said first means are provided within the receiver circuits of said interrogator to present said circuits for evaluating range with pulses having decreased rise time to facilitate greater accuracy of range determination.

4. A DME system according to any preceding claim, in which said interrogator provides circuits functionally interconnected and controlled by microprocessor control means and comprises: clock generator means effective to time said microprocessor control means; pulse pair generating means timed by said clock generator means, said pulse pair generator, triggered by said microprocessor control means, being effective to generate pairs of interrogation pulses; pulse shaping means receiving said pairs of interrogation pulses, with each pulse in each said pulse pair being shaped in a wave-form composed of two edges each with a substantially exponential envelope; programmed synthesizer means, effective to generate a carrier frequency for said interrogation pulses from said pulse shaping means;; amplifier means receiving the signal from said synthesizer means and from said pulse shaping means; amplifier means constructed to improve the noise figure of the receiver said interrogator; intermediate frequency amplifier means and intermediate frequency filtering means effective to reduced the error due to multi-path echos and to present to the demodulator means a wave-form with substantially rectangular envelope; channel detector means with input connected to the output of said intermediate frequency amplifier means; comparator means receiving the output signal of said intermediate frequency amplifier means;; window circuit means, receiving on a first input the output of said comparator means, and on a second input the output of said channel detector means and, on a third and fourth input thereof, the time enabling provided by said microprocessor control means and by counter means respectively; decoder means receiving on a first input the output of said window circuit means, on a second input an output of said synthesizer programming means, and, on a third input, an output of said microprocessor control means, decoder means being provided to recognize whether or not the pair of input pulses from said window circuit means has the correct time spacing to stop said counter means and signal to said microprocessor means that a measurement cycle is terminated.

5. A DME system according to any preceding claim, in which said transponder comprises: clock generator means connected to time a microprocessor control means; pulse pair generator means, timed by said clock generator means, said pulse pair generator means, triggered by said microprocessor control means, being constructed and connected to generate couples of interrogation pulses; pulse shaper means receiving said pairs of interrogation pulses and shaping each pulse in each pair into a wave-form composed of two edges each with a substantially exponential envelope; synthesizer means effective to generate the carrier frequency for said interrogation pulses from said pulse shaper means, said synthesizer means being adapted to be programmed by means of operator controlled digital switching means;; amplifier means receiving the signals from said synthesizer means and from said pulse shaper means: amplifier means having bandwidth characteristics for improving the noise figure of the receiver of said interrogator; intermediate frequency amplifier means and intermediate frequency filtering means effective to reduce the error due to multi-path echo and to present to demodulator means, a wave-form with substantially rectangular envelope; channel detection means having an input connected to the output of said intermediate frequency amplifier means; and comparator means receiving the signal from said intermediate frequency means.

6. A DME system according to claim 5, in which said transponder includes decoding means receiving, on a first input, the output of said comparator means, on a second input the output of all channel detection means, and, on a third input, an output of all clock generating means, the output of all decoding means being connected to dead time generating means, said decoding means being effective to recognize if the input pair of pulses from said comparator means has the correct time spacing.

7. A DME system according to any preceding claim characterized in that said interrogator includes a numerical filtering for reducing the error due to noise.

8. A DME system according to claim 7 characterized in that said numerical filtering includes a Kalman filter with bidimensional state vector.

9. A DME system according to claim 7 characterized in that said numerical filtering is carried out by said microprocessor control means.

1 0. System according to claim 4 or 5, characterized in that said intermediate frequency filter means are produced with surface acoustic wave (SAW) structure or with helical electromagnetic wave structure.

11. A DME system substantially as described with reference to Figs. 3b-1 8 of the accompanying drawings.