WO2019220066A2

WO2019220066A2 - Angle error tracking on the basis of correlation by signal regeneration

Info

Publication number: WO2019220066A2
Application number: PCT/FR2019/051135
Authority: WO
Inventors: Marc THIRY; Hervé Guillon
Original assignee: Centre National d'Études Spatiales
Priority date: 2018-05-18
Filing date: 2019-05-17
Publication date: 2019-11-21
Also published as: FR3081229A1; WO2019220066A3; FR3081229B1

Abstract

The invention relates to a method for estimating misalignment (δθ) on the basis of a sum-channel modulated radiofrequency signal (Σ) and a difference-channel modulated radiofrequency signal (Δ), called primary signals, said method comprising the following steps: - (E1) receiving the primary signals (Σ, Δ) modulated over a central frequency (f0), - (E2) generating a reply signal from the received sum-channel signal (Σ), - (E3) calculating, for every primary signal, at least one intercorrelation product from said primary signal (Σ, Δ) and the reply signal (Σs), - (E4) estimating the misalignment (δθ) by means of the calculated intercorrelation products previously obtained for the sum channel (Σ) and the difference channel (Δ).

Description

The invention relates to a method and a device for estimating misalignment as well as a deviation tracking method and device implementing the estimate of misalignment.

The invention relates in particular to the units, systems and associated methods for guaranteeing the pointing of an antenna by estimating the misalignment of a beam of said antenna with respect to the direction of arrival of the radio-frequency signal to be received.

The deviation tracking uses radio frequency signals whose characteristics are related to the misalignment of the antenna.

STATE OF THE ART

Conventionally the offset of an antenna, denoted dq, is estimated by means of the relative levels of the signals received on the channels å and D (ratio of the amplitudes) of the receiver. Such misalignment is used in a deviation tracking receiver.

Several types of deviation tracking receiver are known. There are completely non-coherent tracking receptors.

This type of receiver analyzes the power received in a band B around the center frequency of the received signal (band B corresponds to the characteristics of the received signal) both on the channel å and on the channel D on reception. Different power levels estimated, we deduce the misalignment of the antenna. This type of receiver works well for a signal-to-noise ratio of the high channel Å allowing the detection of a weak signal on the D-channel. Such a receiver does not need adjustment. There are tracking receivers with consistent detection of the error signal. This type of receiver can operate in coherent or non-coherent signal acquisition mode. In non-coherent mode: the signal of the channel å is detected and continued in frequency (CAF - Automatic Frequency Control) inconsistently by analysis of the power received in a band B around the center frequency of the received signal (band B corresponds to the characteristics of the received signal), and this same signal å standardized by AGC (Automatic Gain Control) is used to detect the signal of the channel D by means of a product in phase of the two signals å and D after the amplitude of the latter has undergone the same normalization as the å way; the fact of proceeding with a product in phase of the two signals å and D supposes that a prior adjustment has been made of the phase difference between the channels of these two signals. This type of receiver requires a signal-to-noise ratio greater than 3 dB (and therefore a symbol energy ratio to noise density, Es / No> 3dB) for various reasons: good operation of the coherent detection, good operation of the AGC (which regulates and normalizes the level according to the received signal, ie the sum Signal + Noise and not only the useful signal), good functioning of the frequency tracking and rejection of the clashes of the loops CAG and CAF on noise.

In coherent mode: a PLL (Phase Lock Loop) is used to track the carrier residue of the channel signal å (if there is one) in phase / frequency in a narrow loop band, and use the reference thus constituted to detect by product in phase the carrier residue of the signal of the channel D after normalization by the coherent AGC of the signals of the two channels. The operating threshold is typically 10 dB of signal-to-noise ratio in the PLL, knowing that the loop band is of the order of 1 kHz. In both cases (acquisition of the coherent or non-coherent signal), the receiver requires a phase adjustment as a function of the reception frequency to guarantee a correct phasing of the signals; moreover this setting is to be taken again in case of replacement of an equipment of the reception chain. This setting is even more critical when the channels DAz and DEI (difference channel for the azimuth axis and the elevation axis respectively) are multiplexed in quadrature: in the latter case, a phase adjustment error is not only by a variation of the deviometry slope but also by the appearance of a crosstalk between the two paths of elevation and azimuth deviation.

In order to overcome the above-mentioned adjustments, correlation tracking receivers are known. One type of this receiver is described in WO2006018571 which operate only for spread spectrum signals. In this document, a correlation between the signal received on the channel lane et and a replica of the spreading code of the signal is carried out; the procedure is the same for channel D and, from the ratio of the complex values of the correlation peaks, the misalignment of the antenna is deduced.

As mentioned this type of receiver does not require any adjustment. However, it is necessary to do a search in frequency fine enough for the correlations to present significant results allowing the deviation tracking.

In addition, this type of receiver does not work for all types of signals.

PRESENTATION OF THE INVENTION

An object of the invention is to estimate the misalignment of an antenna by means of a correlation deviation tracking receiver which operates with all types of signals.

Another object of the invention is to simplify the settings useful for the correlations of this type of receiver. To achieve these aims, the invention proposes a method for estimating misalignment (dq) from a modulated radio frequency signal of sum channel (å) and of a modulated radio frequency signal of difference channel (D), called primary signals. the method comprising the following steps:

- (El) reception of the primary signals (å, D) modulated on a central frequency (fO),

- (E2) generating a replica signal from the received channel signal sum (å),

- (E3) calculating at least one intercorrelation product, for each primary signal, from said primary signal (å, D) and the replica signal (ås),

- (E4) estimation of the misalignment (dq) by means of the previously calculated cross-correlation products obtained for the sum (å) and difference (D) channels.

The cross-correlation product is carried out on a time window T.

In a first embodiment, the step (E2) of generating the replica signal comprises steps of:

- (E21) demodulation of the sum channel signal (å);

- (E22) estimation of the central frequency (fO) from the demodulated signal;

- (E23) decoding the demodulated signal to extract information data;

- (E24) recoding said information data.

Complementarily, in this first embodiment, the step (E2) comprises, after the steps (E21) to (E24), a remodulation step (E25) on the central frequency (fO) to obtain said replica signal, l step (E3) being implemented for the primary signals and the replica signal. In a second embodiment, step (E2) comprises, after steps (E21) to (E24), a step of generating a baseband complex signal to obtain said replica signal, the method further comprising a step (E3 ') of frequency descent of the primary signals on the estimated central frequency (fO), the step (E3) being implemented for said primary signals down in frequency and said replica signal.

The method, for both embodiments, may comprise a step (Es) of storing the primary signals to delay them before the implementation of step (E3).

In a particular variant, the primary signals are digital, and the storage step is preferably a bufferization. The storage step (Es) is configured to delay the primary signals by a time corresponding to the implementation of the step (E2). The storage step (Es) can be of the order of ten to several hundreds of milliseconds.

The method may further comprise a step (EN) of digitizing the primary signals. This step is done after the reception step E1, before the subsequent processing, and therefore in particular before the step E2 and the storage step Es.

To calculate the misalignment, step (E4) may comprise the following steps:

calculation of the values of the peaks CA, CA for the sum channel and for the difference channel,

calculating the misalignment dq by the formula:

where sign () is the function that assigns 1 or -1 depending on the positive or negative sign and Re () designates the real part function. The invention also proposes a method of differential tracking of an antenna, implemented subsequently to a depointing estimation method as described above, and comprising a step (E5) for generating a displacement instruction (C50). of the antenna based on the estimated misalignment (dq), preferably on an azimuthal axis and on an elevation axis.

The invention also proposes a deviation tracking unit, comprising a processor, a buffer memory, a displacement instruction generating module, the unit being configured to implement a depointing method or a deviation tracking method as described. The invention also proposes a deviation tracking system comprising a telemetric system comprising an antenna configured to receive radio frequency signals and an antenna source for generating a sum channel signal (å) and a difference channel signal (D). said block further comprising a deviation tracking unit as described above, in which the direction of the antenna (A) is controllable by the movement instructions generated by the unit.

The advantages of the invention are twofold for a deviation receiver operating on these principles:

it can operate with a lower signal-to-noise ratio than at present for signals without carrier residue:

o Phase demodulation allows the acquisition of the signal at low ES / No (<3dB) and the application of a narrow narrow band AGC on both channels.

o The correlation using a noiseless replica for the sum channel is by design more accurate than the classical cross correlation there is no phase adjustment to be made, since the principle of the intercorrelation of the replica to the sum and delta signals over a given time range makes it possible to overcome it. This relieves maintenance operations in a very important way.

PRESENTATION OF FIGURES

Other characteristics, objects and advantages of the invention will emerge from the description which follows, which is purely illustrative and nonlimiting, and which should be read with reference to the appended drawings, in which:

- Figure 1 illustrates the hardware for implementing the method.

FIG. 2 illustrates a first embodiment of the invention; FIG. 3 illustrates a second embodiment of the invention.

DETAILED DESCRIPTION

FIG. 1 illustrates a deviation tracking device according to one embodiment of the invention.

The deviation tracking device comprises a receiving antenna A, a deviation tracking unit 1, an antenna orientation device A1, a telemetric system 2.

The misalignment exists for the axes Az and El, as indicated in the introduction. For the sake of simplicity, only obtaining a single misalignment dq will be described. The antenna A is configured to receive a radio frequency signal from, for example, a satellite S, a launcher or any other vehicle. This signal to be received is considered to be modulated around a frequency f0. The telemetric system 2 receives this modulated signal from the machine observed by the antenna A. This received signal is processed by an antenna source 21 to provide a channel signal sum å and a difference channel signal D whose amplitude relative to that of the sum signal å is proportional to misalignment for low misalignments. The antenna source 21 is also capable of digitizing analog signals. These signals (analog or digital) are then optionally amplified. It is assumed for this that the antenna A is originally correctly oriented towards the satellite to be observed to allow the acquisition of the signal (typically: in the lobe at 3dB of the antenna).

These sum å and D difference channel signals are then sent to and received by the deviation tracking unit 1 which proceeds to their processing. These signals will be called primary signals later.

The antenna orientation device Al is configured to orient the antenna mechanically (motor, etc.) or digital for the active antennas.

The deviation tracking unit 1 comprises a processor 11 (or any other equipment capable of fulfilling the same function) for implementing a method of estimating the offset dq of the antenna as a function of the channel and train signals. difference D on the one hand and implement a deviation tracking method from the estimated misalignment.

In particular, the deviation tracking unit is thus able to generate an estimate of the misalignment dq and to provide an orientation setpoint Cd0 of the antenna so that it is oriented correctly.

This orientation instruction Cd0 is supplied to the device A1 for orienting the antenna A. The deviation tracking unit 1 also comprises a buffer memory 12 capable of either buffering data, that is to say of delaying their transmission, and, if necessary, a non-volatile memory 13. The memory buffer 12 is preferably a circular buffer.

The deviation tracking unit 1 further comprises an input interface 15 for receiving data (coaxial, wired or wireless) and an output interface 16 for transmitting the processed data. The input interface 15 thus comprises at least two ports (physical or virtual) for respectively a channel signal sum å and a difference channel signal D (three ports if we consider the signal channel sum å and two difference channel signals DAz and DEI).

A method of estimating the dq misalignment implemented by a processor of the deviation tracking unit 1 is described below with reference to FIG.

The sum å and difference channel signal D are received (step E1) by the deviation tracking unit 1 in order to calculate the misalignment dq, in particular by calculating a replica signal, generated from the sum channel signal, removed from the noise by a conventional demodulation / decoding / synchronization method followed by a conventional coding / modulation process to correlate with the primary signals. These primary signals are modulated around a central frequency f0.

Just after the step E1, a step EN of digitizing the primary signals is advantageously implemented, before the treatments which will be described later. This allows working with digital signals.

In a step E2, a local replica signal is generated from the received channel signal sum. The generation of this replica signal notably comprises steps of demodulation (step E21) of the channel signal sum å to estimate (step E22) the central frequency f0.

Then, this demodulated signal is decoded (step E23) to obtain information data. This information data can be provided to a user as output data. From the information data thus obtained, a recoding of the data is performed (step E24) in order to obtain a replica signal at the same frequency as the primary signals (step E25 in FIG. 2) or a complex signal in a band of base (step E25 'in Figure 3). Then, in a step E3, a correlation product is implemented on a time window of duration T (that is to say a given time interval T), carried out in particular by the processor of the unit 11 of the unit 1. For each primary signal, i.e., the sum channel signal and the difference channel signal D, an intercorrelation product calculation is made from said primary signal and the replica signal.

The cross-correlation products make it possible to identify peaks CA, CA, respectively resulting from the correlation between the replica signal resulting from the sum channel signal and each primary signal.

The misalignment dq can be given according to the following formula:

where sign () is the function that assigns 1 or -1 depending on the positive or negative sign, and Re () designates the real part function.

In other words, the ratio of the modules of these peak values C, CA is proportional to the misalignment of the antenna A during the tracking. angular deviation. The direction of misalignment is determined by the relative sign of the real parts of Cå and CA.

There may be other calculations or alternatives to get the misalignment, but the formula above gives good results.

The value of the peak C is generally easy to obtain; on the other hand, the value of the peak C D on the difference channel D can be more difficult to estimate, in particular when the antenna A is well pointed (and thus the signal on the difference channel D is weak, or even null, with the noise close). However, the processing delays between the sum å and difference D channels can be known, which can make it possible to know a priori the position of the correlation peak CA on the difference channel A with respect to the position of the correlation peak C sur on the track sum å. By limiting the search for the correlation peak CA on the difference correlation path D in this time zone, it is possible to improve the pointing performance (reduction of the calculation duration and therefore decrease of the delays introduced by the receiver and / or increase of the duration of integration of measurements, on the other hand, it limits the search area of a low correlation peak, the risk of identifying a peak due to noise is reduced).

To implement the correlations of the step E3, a step Es of storage of the primary signals å, D is performed in order to delay them for a duration corresponding substantially to the processing time to generate the non-noisy replica signal å before the correlation. More details will be given for each of the embodiments of steps E2 and E3.

Since the data processed by the deviation tracking unit 1 are digital, the storage step Es corresponds to a buffering, preferably circular, that is to say that the first data entered will be the first data output (FIFO ). Given the time scales involved in the signal processing, related to the current calculation capabilities, the storage step Es lasts between ten and several hundreds of milliseconds. After the demodulation step E21, a decoding step E23 is generally performed to obtain the information conveyed by the sum signal and transfer it to the users concerned (whether an operator or other data processing blocks, not concerned). by the invention). We will talk about "user data". Subsequent to this decoding step E23, a coding step E24 (recoding more specifically) is necessary.

The coding algorithms are known and conventionally used: turbo-code decoding, convolutional, Reed Solomon, BDv-S2 frames, etc.

Two embodiments are described below with respect to the generation of the replica signal and its use. First embodiment: intercorrelation at fO (FIG. 2)

At the end of the demodulation step (step E21) and obtaining the central frequency f0 (step E22), and in addition to the decoding steps Ed / recoding Ec (steps E23 and E24), the demodulated signal (and in practice decoded then recoded) is again modulated using the central frequency fO. More precisely, we are talking about remodulation (step E25). A non-noisy digital replica of the sum channel signal is then available (assuming decoding has corrected all errors).

In this first embodiment, the replica signal therefore corresponds to this resodulated sum channel signal. It will serve as a reference signal. The cross-correlation product calculation step E3 thus comprises the computation of a product of intercorrelation between the track signal sum å, which has been stored, and the sum signal remodulated (called the replica signal) and a product of cross-correlation between the difference channel signal D and the remodulated sum channel signal (said replica signal).

Since steps E21 to E25 take time, the storage duration is determined to correspond to the processing time of these steps.

Second Embodiment: Baseband Intercorrelation (FIG.

3) At the end of the demodulation step (step E21) and obtaining the central frequency f0 (step E22), and in addition to the decoding step Ed / recoding Ec, the demodulated, decoded and then recoded signal n ' this time is remodulated. A complex baseband signal E25 'is then generated from the decoded signal during step E23. Whereas in the case of FIG. 2, the signal generated at the end of step E25 would have been of the form:

it will be here, at the end of the step E25 'of the form: s (t) = A (t). exp (jp (t))

In parallel, the sum and difference channel signals Δ, D are each lowered E3 'in baseband frequency using the estimated frequency f0; at the end of this frequency descent, these signals are considered as baseband (that is to say centered around the zero frequency) The cross-correlation product calculation step E3 therefore comprises the calculation of a cross-correlation product between the sum-of-channel signal, which has been stored and then downgraded and the complex replica signal generated and a product of intercorrelation between the difference channel signal D and the complex replica signal generated.

Since the steps E21 to E25 'take time, on a computer scale, the duration of the storage is determined to correspond to the processing time of these steps, minus the time necessary to effect the frequency descent on the primary signals ( In practice, this is a very fast operation - multiplication and low-pass filtering.

Complex signal generation may appear as a more delicate and difficult method to implement. However, in case of simple modulation, such as BPSK, QPSK or OQPSK, the complex symbols are simple to handle (± 1 for BPSK, ± 1 ± j for QPSK and OQPSK).

Similarly to the first embodiment, a noiseless replica of a primary signal is used here.

As indicated above, the misalignment is advantageously used in a deviation tracking method. In particular, in the context of this method, the calculated misalignment dq is used to generate a displacement setpoint (generally an angular setpoint) which delivers a setpoint C50 typically in the form of a deviation measurement voltage K × dq, K being a deviation measurement voltage coefficient.

As indicated above, the method is advantageously implemented for azimuth axes Az and El elevation.

Claims

claims

A method for estimating misalignment (dq) from a modulated radio signal modulated channel (å) and a modulated difference channel radio signal (D), called primary signals, the method comprising the following steps:

- (E2) generating a replica signal from the received channel signal sum (å),

The method of claim 1, wherein the step (E2) of generating the replica signal comprises steps of:

- (E21) demodulation of the sum channel signal (å);

- (E22) estimation of the central frequency (fO) from the demodulated signal;

- (E23) decoding the demodulated signal to extract information data;

- (E24) recoding said information data.

3. The method according to claim 2, wherein the step (E2) comprises, after the steps (E21) to (E24), a remodulation step (E25) on the central frequency (fO) to obtain said replica signal, the step (E3) being implemented for the primary signals and the replica signal.

The method according to claim 1 or 2, wherein step (E2) comprises, after steps (E21) to (E24), a step of generating a complex baseband signal to obtain said replica signal. the a method further comprising a step (E3 ') of frequency descent of the primary signals on the estimated center frequency (fO), the step (E3) being implemented for said frequency-downed primary signals and said replica signal.

5. Method according to any one of the preceding claims, comprising a step (Es) for storing the primary signals to delay them before the implementation of step (E3). The method of claim 5, wherein the primary signals are digital, and the storage step is preferably a bufferization.

7. The method of claim 5 or 6, wherein the storage step (Es) is configured to delay the primary signals by a time corresponding to the implementation of step (E2).

8. Method according to any one of claims 5 to 7, wherein the storage step (Es) is of the order of ten to several hundred milliseconds.

9. Method according to any one of the preceding claims, comprising a step (EN) for digitizing the primary signals, performed before step (E2).

The method of any one of the preceding claims, wherein the step (E4) comprises the following steps:

calculating the misalignment dq by the formula:

where sign () is the function that assigns 1 or -1 depending on the positive or negative sign and Re () designates the real part function.

11. Differential tracking method of an antenna, implemented after a misalignment estimation method according to any one of the preceding claims, and comprising a step (E5) for generating a displacement instruction (C50). of the antenna

(21) based on the estimated misalignment (dq), preferably on an azimuthal axis and on an elevation axis.

12. Differential tracking unit (1), comprising a processor (11), a buffer memory (12), a displacement instruction generating module (14), the unit (1) being configured to implement a method misalignment estimation system according to any one of claims 1 to 10. 13. Deviation tracking system comprising a telemetry system (2) comprising an antenna (A) configured to receive radio frequency signals and an antenna source (21) for generating a sum channel signal (å) and a difference channel signal (D), said block further comprising a deviation tracking unit (1) according to claim 12, wherein the direction of the antenna (A) is controllable by the displacement instructions generated by the deviation tracking unit (1).