FR3147470A1 - Validation of the payload of a telecommunications satellite based on a calculation of the radiated power ratio between useful signal and intermodulation noise - Google Patents

Abstract

The invention relates to a method (100) for monitoring the performance, in terms of radiated noise power, of a payload of a telecommunications satellite based on an active multi-beam antenna. The method comprises in particular the determination (105) of a spectral density of intermodulation noise power radiated to the ground by all the sources of the antenna from: the carrier allocation plan and the associated RF powers making it possible to fulfill the telecommunications mission of the satellite, the weighting coefficients of the beamforming network making it possible to form the radio beams responding to the mission, the linearity characteristics of the amplifiers associated with the different sources of the antenna, and the radiation patterns of the different sources. The method makes it possible to determine (109) a radiated power ratio between useful signal and intermodulation noise ("radiated C/IM") for at least one carrier and for at least one geographical area of interest. Figure for the abstract: Fig. 4

Description

Validation of the payload of a telecommunications satellite based on a calculation of the radiated power ratio between useful signal and intermodulation noise Field of invention

The present invention relates to the field of performance control of the payload of a telecommunications satellite. More particularly, the invention relates to a method for validating the configuration of the payload of a telecommunications satellite based on an estimation of a radiated power ratio between useful signal and intermodulation noise (“radiated C/IM”).

State of the art

Some telecommunications satellites are used to transmit information to fixed reception points, which is known as "fixed satellite service" (FSS) or DTH broadcasting (English acronym for "Direct-To-Home"). This is the case, for example, for broadcasting television programs or for international telephony.

Other telecommunications satellites are assigned to high-speed Internet connection services corresponding to HTS type missions (acronym for “High-Throughput Satellite”).

For the same amount of allocated radio spectrum, a satellite responding to an HTS mission provides significantly more throughput than a satellite responding to an FSS mission. This significant increase in throughput for HTS missions is achieved by reusing frequency channels in tightly focused radio beams arranged to cover a geographic region of interest. The increase in throughput is also accompanied by an increase in coverage flexibility (beam variations over time).

Satellites responding to an HTS type mission generally carry an active multibeam antenna. Such an antenna comprises a beam-forming network (BFN) connected to a network of sources. Phase and amplitude weighting coefficients are associated respectively with the different sources and the different frequency channels in order to form the desired radio beams. Each source is associated with an amplifier adapted to amplify the signal to be emitted by the source.

The payload of a telecommunications satellite can be controlled and updated by a mission control center (MCC), in particular to respond to changes in an ongoing mission, or even to respond to a new mission significantly different from the current mission.

To monitor the payload performance of a telecommunications satellite, it is necessary to monitor the radio link budget between the satellite and a receiving station located in a geographical area of interest on the Earth's surface.

The non-linearity of the amplifiers of the active multibeam antenna generates intermodulation noise. The radiated power ratio between the useful signal and the intermodulation noise (“C/IM ratio”) is an important element contributing to the calculation of the radio link budget of a communication between a satellite and a ground station.

However, there is currently no satisfactory solution for quickly calculating a C/IM ratio for an active multibeam antenna comprising a large number of sources (several tens or even several hundreds of sources) and configured to form a large number of beams (several hundreds or even several thousands of beams).

The paper “Calculation of third order intermodulation spectrum”, Y.H. LAU et al., describes a method for fast calculation of an intermodulation noise spectral density in a conducted manner (and not in a radiated manner).

The present invention aims to remedy all or part of the drawbacks of the prior art, in particular those set out above.

• une détermination d'un « masque normalisé » de densité spectrale de puissance de bruit d'intermodulation à partir des canaux fréquentiels composant les signaux d'entrée et des niveaux de puissance respectifs desdits canaux fréquentiels,
• une détermination d'un point d'opération correspondant à une puissance de sortie de l'amplificateur, le point d'opération étant déterminé en fonction des niveaux de puissance des canaux fréquentiels des signaux d'entrée, et en fonction des coefficients de pondération associés audit amplificateur,
• une détermination d'une puissance totale de bruit d'intermodulation en fonction du point d'opération et de caractéristiques de linéarité de l'amplificateur,
• une détermination d'un « masque absolu » de densité spectrale de puissance de bruit d'intermodulation par une pondération du masque normalisé avec la puissance totale de bruit d'intermodulation.

For this purpose, and according to a first aspect, the present invention proposes a method implemented by a computer for validating a configuration of a payload of a telecommunications satellite. The satellite is configured to transmit a signal with an active multi-beam antenna comprising a beamforming network. The beamforming network is connected to a plurality of sources and adapted to simultaneously form several beams of interest serving geographical areas on the surface of the Earth. The beamforming network is configured to receive as input, for each beam of interest to be formed, an input signal comprising one or more frequency channels with respective power levels, and to transmit to each source a linear combination of the input signals weighted by weighting coefficients, each source being connected to an amplifier. The method comprises for each amplifier:

• a determination of a “normalized mask” of intermodulation noise power spectral density from the frequency channels composing the input signals and the respective power levels of said frequency channels,
• a determination of an operating point corresponding to an output power of the amplifier, the operating point being determined as a function of the power levels of the frequency channels of the input signals, and as a function of the weighting coefficients associated with said amplifier,
• a determination of a total intermodulation noise power as a function of the operating point and linearity characteristics of the amplifier,
• a determination of an “absolute mask” of intermodulation noise power spectral density by weighting the normalized mask with the total intermodulation noise power.

• une détermination d'un « masque rayonné » de densité spectrale de puissance de bruit d'intermodulation à partir du masque absolu de densité spectrale de puissance de bruit d'intermodulation de l'amplificateur auquel la source est reliée, et à partir d'un diagramme de rayonnement de la source.

The method includes for each source:

• a determination of a "radiated mask" of intermodulation noise power spectral density from the absolute intermodulation noise power spectral density mask of the amplifier to which the source is connected, and from a radiation pattern of the source.

• une détermination d'un « masque rayonné total » de densité spectrale de puissance de bruit d'intermodulation par une sommation en puissance des masques rayonnés des différentes sources.

The method further includes:

• a determination of a “total radiated mask” of intermodulation noise power spectral density by a power summation of the radiated masks of the different sources.

• une détermination d'une puissance rayonnée de bruit d'intermodulation par une intégration, sur une sous-bande de fréquences correspondant audit canal fréquentiel, du masque rayonné total de densité spectrale de puissance de bruit d'intermodulation pour la zone géographique d'intérêt,
• une détermination, pour le canal fréquentiel d'intérêt, d'une puissance isotrope équivalente rayonnée par l'antenne active multifaisceaux,
• une détermination d'un rapport « C/IM rayonné » entre la puissance isotrope équivalente rayonnée et la puissance rayonnée de bruit d'intermodulation,
• une vérification d’un critère prédéterminé en fonction du rapport « C/IM rayonné » ainsi déterminé.

For at least one frequency channel of interest, and for at least one geographic area of interest:

• a determination of a radiated intermodulation noise power by an integration, over a frequency sub-band corresponding to said frequency channel, of the total radiated intermodulation noise power spectral density mask for the geographical area of interest,
• a determination, for the frequency channel of interest, of an equivalent isotropic power radiated by the active multibeam antenna,
• a determination of a “radiated C/IM” ratio between the equivalent isotropic radiated power and the radiated intermodulation noise power,
• a verification of a predetermined criterion based on the “radiated C/IM” ratio thus determined.

The method according to the invention thus provides a rapid and efficient method for an analytical evaluation of the performance of the radiated power ratio between useful signal and intermodulation noise. A configuration of the payload of a telecommunications satellite can thus be validated as satisfying a specification. Several configurations can advantageously be tested and compared to select the configuration that best meets a need.

• le plan de porteuses de la mission de télécommunication considérée (les canaux fréquentiels composant les signaux d’entrée du réseau de formation de faisceaux),
• les poids utilisés par le réseau de formation de faisceaux de l’antenne active (les coefficients de pondération associés aux différents canaux fréquentiels et aux différentes sources),
• les caractéristiques de linéarité des amplificateurs associés respectivement aux différentes sources de l’antenne active, et
• les diagrammes de rayonnement des différentes sources de l’antenne active.

The proposed solution advantageously allows the calculation of a ground-radiated performance per carrier by taking into account:

• the carrier plan of the telecommunications mission considered (the frequency channels composing the input signals of the beamforming network),
• the weights used by the active antenna beamforming network (the weighting coefficients associated with the different frequency channels and the different sources),
• the linearity characteristics of the amplifiers associated respectively with the different sources of the active antenna, and
• the radiation patterns of the different sources of the active antenna.

The method according to the invention is applicable to any type of active multi-beam antenna. It may be a transmitting antenna or a receiving antenna. The antenna may comprise one or more beamforming networks of the analog or digital type. It may also be an antenna with or without a reflector. The active antenna may have several reflectors and/or several source networks. The amplifiers associated with the different sources may be of the SSPA type (acronym for “Solid State Power Amplifier”) or tube-based.

Advantageously, the execution time of the method remains very short and little impacted by the number of frequency channels, the number of sources and the number of beams formed.

The method according to the invention can be applied to tools for designing, parameterizing or validating the configuration of the payload of a telecommunications satellite, or to tools for modeling the performance of the payload of a telecommunications satellite.

The method can be implemented for satellites operating in different types of orbit: geostationary orbit (GEO), medium Earth orbit (MEO), low Earth orbit (LEO), Molnia orbit, etc.

In particular embodiments, the invention may further comprise one or more of the following features, taken individually or in any technically possible combination.

In particular implementations, the normalized intermodulation noise power spectral density mask is common to all amplifiers.

In particular embodiments, the normalized intermodulation noise power spectral density mask is determined for an amplifier based on the weighting coefficients associated with said amplifier.

In particular embodiments, the determination of the normalized intermodulation noise power spectral density mask involves a calculation of third-order intermodulation products generated by the frequency channels of the input signals.

In particular embodiments, the determination of the normalized intermodulation noise power spectral density mask further comprises a calculation of intermodulation products of odd order strictly greater than three generated by the frequency channels of the input signals.

In particular embodiments, the determination of the normalized intermodulation noise power spectral density mask involves the application of a correction value aimed at compensating for intermodulation products of order strictly higher than three not taken into account.

In particular embodiments, the determination of the normalized intermodulation noise power spectral density mask comprises a normalization calculation based on the result of an integration, over a frequency band, of the intermodulation noise power spectral density.

In particular embodiments, the determination of the normalized intermodulation noise power spectral density mask comprises a normalization calculation based on the result of an integration, over a frequency band, of the intermodulation noise power spectral density for power values within a predetermined range of values.

In particular embodiments, the method further comprises applying a correction value to the “radiated C/IM” ratio. The correction value is determined based on a frequency channel reuse factor for forming the beams of interest.

In particular embodiments, the frequency band is one of the bands W, Q, V, Ka, Ku, L, S, C.

According to a second aspect, the present invention relates to a computer program product comprising instructions which, when executed by a computer, cause the latter to implement the steps of a method for validating a configuration of a payload of a telecommunications satellite according to any of the preceding implementation modes.

According to a third aspect, the present invention relates to a computer-readable recording medium comprising instructions which, when executed by a computer, cause the latter to implement the steps of a method of validating a configuration of a payload of a telecommunications satellite according to any of the preceding embodiments.

Presentation of figures

The invention will be better understood by reading the following description, given as a non-limiting example, and made with reference to the following figures:

a schematic representation of a satellite telecommunications system,

a schematic representation of an active multibeam antenna of a telecommunications satellite,

a schematic representation of a computer configured to implement a method according to the invention for validating the configuration of the payload of a telecommunications satellite,

a schematic representation of the main steps of a first mode of implementation of the method according to the invention,

a schematic representation of the main steps of a second mode of implementation of the method according to the invention.

In these figures, identical references from one figure to another designate identical or similar elements. For the sake of clarity, the elements shown are not necessarily to the same scale, unless otherwise indicated.

Detailed description of an embodiment of the invention

There schematically represents a satellite telecommunications system. As illustrated in the , the system comprises at least one telecommunications satellite 30 in orbit around the Earth 50. In the example considered, the mission of the telecommunications satellite 30 corresponds to a high-speed Internet connection service (HTS type service). The case is considered in a non-limiting manner in which the satellite 30 is in geostationary orbit (GEO type orbit). However, nothing excludes, according to other examples, considering a non-geostationary Earth orbit, such as a low-altitude orbit (LEO type orbit) or a medium-altitude orbit (MEO type orbit).

Conventionally, the telecommunications satellite 30 can allow data exchanges between two gateway stations 40 via radio communication links 41 established respectively between each gateway station 40 and the telecommunications satellite 30. According to another example, the telecommunications satellite 30 can allow data exchanges between a user 42 and a gateway station 40 via on the one hand a radio communication link 41 established between the satellite 30 and the user 42 and on the other hand a radio communication link 43 established between the satellite 30 and the gateway station 40. According to yet another example, the telecommunications satellite 30 can allow data exchanges between two users 42 via radio communication links 43 established respectively between each user 42 and the telecommunications satellite 30.

Each radio communication link 41, 43 can support a downward direction (“downlink” in the English literature) for the transmission of data by the satellite 30 and/or an upward direction (“uplink” in the English literature) for the reception of data by the satellite 30.

The transmissions and/or receptions on the radio communication links 41, 43 are carried out, for example, in one or more frequency bands among the bands W, Q, V, Ka, Ku, L, S, C, etc.

In the example considered and illustrated on the , the telecommunications satellite 30 is controlled by a mission control center 17 located on the ground, on the surface of the Earth 50. The mission control center 17 makes it possible in particular to update the payload of the telecommunications satellite 30 to respond to possible changes in the current mission, or even to respond to a new mission significantly different from the current mission.

Conventionally, the payload of the telecommunications satellite 30 can be configured remotely by sending commands through radio signals transmitted on a control link 16 established between the telecommunications satellite 30 and the mission control center 17.

As illustrated in the , the mission control center 17 has an antenna 15 for communicating with the telecommunications satellite 30 on the control link 16.

The mission control center 17 also includes a computer 10 configured to implement a method for validating the payload of the telecommunications satellite 30. It is indeed interesting to be able to control the performance of the payload of the telecommunications satellite 30 in terms of radio link budget between the satellite 30 and a receiving station (corresponding for example to a gateway station 40 or to a user 42) located in a geographical area of interest on the surface of the Earth 50. More particularly, it is interesting to be able to evaluate, for a given carrier and for a given geographical area of interest, a radiated power ratio between useful signal and intermodulation noise (“C/IM ratio”). The C/IM ratio is indeed an important element in the radio link budget of a communication established between a satellite and a terrestrial receiving station.

To meet the needs of its HTS mission, particularly in terms of throughput flexibility and geographic coverage, the 30 telecommunications satellite carries a 20 active multi-beam antenna.

There schematically represents such an active multi-beam antenna 20. The antenna 20 comprises in particular a beamforming network 25 connected to a network 23 of sources 24. The method according to the invention is particularly well suited to the case where the number of sources 24 is large, for example at least fifty (M ≥ 50), or even at least one hundred (M ≥ 100). However, nothing would prevent the invention from being applied to cases where the number of sources would be smaller. In the example considered, the network 23 of sources comprises two hundred sources 24 (M = 200).

In the example considered and illustrated on the , the antenna 20 also comprises a reflector 22. The array 23 of sources 24 is for example positioned at the focal point of the reflector 22 (AFR type antenna, acronym for “Array Fed Reflector”). According to another example, the array 23 of sources 24 is offset relative to a focal point of the reflector 22 (DAFR type antenna, acronym for “Defocused Array Fed Reflector”). However, nothing precludes using an antenna 20 without a reflector (DRA type antenna, acronym for “Direct Radiating Array”), or an antenna comprising several reflectors. Nothing would also prevent considering an antenna 20 with several reflectors.

The beamforming network 25 may be “adaptive,” meaning that the beams formed by the beamforming network 25 may vary over time.

The beamforming network 25 may be analog or digital. An analog beamforming network performs processing on analog signals (electrical or microwave), for example with discrete electronic circuits such as attenuators, phase shifters or time delay circuits. A digital beamforming network performs processing on digital signals.

The beamforming network 25 may be of any type known to those skilled in the art, including phase shifter type or true time delay type.

In the remainder of the description, we consider the case where the antenna 20 is a transmitting antenna using an analog adaptive beamforming network 25 of the phase shifter type.

As illustrated in the , the beamforming network 25 receives as input a plurality of input signals each comprising one or more frequency channels with respective power levels. In the example considered and illustrated in the , we consider M input signals. Each input signal corresponds to a beam of interest to be formed. There may be several hundred, or even several thousand beams to be formed simultaneously. The method according to the invention is particularly well suited to the case where the number of beams to be generated is large, for example at least equal to five hundred (L ≥ 500), or at least equal to one thousand (L ≥ 1000). However, nothing would prevent the invention from being applied to cases where the number of beams would be smaller. As a non-limiting example, we will take the case where there are two thousand beams to be formed (L = 2000). These input signals correspond, for example, to a carrier plan defined by an operator of the telecommunications satellite 30. Each carrier corresponds to a frequency channel occupying a sub-band Δf of a frequency band ΔF. The frequency band ΔF corresponds for example to the Ka band which extends between 27.5 and 31 GHz for transmission communications. The sub-band Δf occupied by a frequency channel has for example a bandwidth of 10 MHz. Each frequency channel is associated with a predetermined power level. A beam can be formed by one or more carriers (one or more frequency channels per input signal). It should be noted that the same frequency channel can be used for different beams (the same frequency channel can be present in different input signals).

The power levels (conducted RF power) of the frequency channels in the input signals can be determined from EIRP (Equivalent Isotropic Radiated Power) specifications provided by the operator. These specifications provide the minimum EIRP level that the antenna must provide for each frequency channel. The EIRP is a combination of the antenna directivity and the conducted RF power. The antenna directivity is known via the radiation patterns of the sources 24 and the weighting coefficients used by the beamforming network 25 (see below). The conducted RF power required to meet the specifications can therefore be deduced.

Conventionally, the beamforming network 25 is configured to apply phase and amplitude weighting coefficients to the different sources 24 and to the different frequency channels in order to form the desired radio beams to serve geographic areas of interest on the surface of the Earth 50.

In other words, the beamforming network 25 is configured to transmit to each source 24 a linear combination of the input signals weighted by the phase and amplitude weighting coefficients in order to form the desired radio beams. The weighting coefficients make it possible to modulate the position and shape of each of the beams to be formed.

For this purpose, the beamforming network 25 may comprise, in a known manner, duplicator circuits, phase shifters and attenuators making it possible to apply an amplitude and phase weighting coefficient to each frequency channel for each source 24 of the network 23.

It should be noted that at any given time, not all sources 24 of the network 23 are necessarily used to form the desired beams. If this is the case, it is sufficient to take into account the sources 24 which are actually used.

Each source 24 is connected to an amplifier. This may be, for example, an SSPA type amplifier or a tube-based amplifier. In the example considered, an amplifier is associated with one and only one source 24 (bijection between the amplifiers and the sources 24). However, nothing would prevent us from considering the case where the same amplifier would be associated with several sources 24.

An amplifier associated with a source 24 has the role of amplifying the power of the signal transmitted to the source 24 by the beamforming network 25 before the radio emission of said signal by the source 24. An amplifier therefore receives as input the linear combination of the weighted input signals intended for the source 24 with which (or the sources 24 with which) the amplifier is associated.

An amplifier generally exhibits nonlinear characteristics when the expected power level at the amplifier output is high. The more power the amplifier must provide at the output, the more nonlinear it is, and the more it will generate intermodulation products.

The intermodulation products resulting from the non-linearity of the amplifiers impact the link budget of a communication established between the telecommunications satellite 30 and a receiving ground station. This impact can be evaluated by calculating a ratio between useful signal and intermodulation noise (“C/IM ratio”) at the level of the signal radiated on the ground in a geographical area of interest (geographic area served by a radio beam emitted by the active multi-beam antenna 20). This radiated C/IM ratio can in particular be calculated for a given carrier (i.e. for a particular frequency channel).

Figures 4 and 5 represent the main steps of two modes of implementation of a method 100 aimed at controlling the configuration of the payload of a telecommunications satellite 30. The method 100 according to the invention makes it possible to calculate a radiated C/IM ratio for at least one carrier and for at least one geographical area of interest. The method 100 then makes it possible to verify a predetermined criterion as a function of the calculated radiated C/IM ratio.

In the example considered, this method is implemented by a computer 10 of the mission control center 17. In other words, in the example considered, this method is used during an operational phase of the telecommunications satellite 30. It should however be noted that this method could also be used during upstream phases, for example during a design or dimensioning phase of the payload of the telecommunications satellite 30.

There schematically represents a computer 10 configured to implement the method 100 according to the invention to validate the configuration of the payload of a telecommunications satellite 30. As illustrated in the , the computer 10 comprises at least one processor 11 and at least one memory 12. The memory 12 stores code instructions of a computer program 13 which, when executed by the processor 11, configures the processor 11 to implement the method 100 according to the invention. The computer 10 may also comprise a user interface 14 (also known as HMI, acronym for “Human-Machine Interface”) to receive possible input parameters. Alternatively or in addition, the computer 10 may also be connected to another machine configured to provide all or part of the input parameters.

As illustrated in FIGS. 3 and 4 , the method 100 takes various parameters as input. A first set of parameters corresponds to the power levels of the frequency channels making up the input signals. This corresponds to a frequency plan allocated for the mission assigned to the telecommunications satellite 30 . This frequency plan is for example provided by an operator of the telecommunications system described with reference to FIG. The frequency plan indicates the RF conducted power level of each frequency channel contained in the input signals.

A second set of parameters corresponds to the weighting coefficients associated with the different frequency channels and the different sources 24. These weighting coefficients include in particular amplitude weights applied by the beamforming network 25 to the different frequency channels for each source 24. These weights correspond to a particular configuration of the beamforming network 25 of the antenna 20 to form radio beams making it possible to respond to the mission of the telecommunications satellite 30 at a given time. In the present application, the expression “the weighting coefficients associated with an amplifier” must be understood as “the weighting coefficients associated with a source 24 (or the sources 24 with which) the amplifier is associated”.

A third set of parameters corresponds to linearity characteristics of the amplifiers associated with the different sources 24. The linearity characteristics of an amplifier are for example defined by a curve describing a noise power ratio (NPR) as a function of an output power of the amplifier (operating point, or OBO for output back-off). This curve NPR = f(OBO) is for example provided by the manufacturer of the amplifier. The OBO corresponds to an output power relative to the saturation point of the amplifier.

A fourth set of parameters corresponds to the radiation patterns of the different sources 24 of the antenna 20. Each radiation pattern is a representation of the distribution in three-dimensional space of a quantity representative of the power radiated by a source 24.

As illustrated in Figures 3 and 4, the method 100 comprises a step 101 of determining a “normalized mask” of power spectral density (PSD) of intermodulation (IM) noise from the power levels of the frequency channels composing the input signals.

The power spectral density of the noise generated by the intermodulation products of the frequency channels contained in the frequency plane can be calculated analytically as described in the document "Calculation of third order intermodulation spectrum", Y.H. LAU et al., ELECTRONICIS LETTERS, October 11th, 1990, Vol. 26, No. 21. This method makes it possible to calculate the power spectral density of the intermodulation products by convolving the power spectrum of the signal present at the input of an amplifier.

It is possible to consider only third-order intermodulation products (these are those which dominate the power spectrum of the intermodulation noise). Third-order intermodulation products include triple products (of the type , formed from three distinct frequency channels centered respectively on the frequencies , And ), and double products (of the type , formed from two distinct frequency channels centered respectively on the frequencies , And ). The power of a triple product is 6 dB louder than that of a double product.

The power spectral density of the noise generated by third-order intermodulation products can then be written in the form:

In this expression, the operator corresponds to a convolution operation. N corresponds to the number of different frequency channels contained in the input signals. The term corresponds to the power spectrum of a frequency channel of index . For example, it is considered to take the form of a slot centered at the frequency and bandwidth Δf. The amplitude of the slot corresponds to the power level assigned to the frequency channel (different power levels are assigned to different frequency channels, depending on the frequency plan considered). We then have ) And . The term corresponds to the power spectrum of the input signal: .

Convolutions involving the power spectrum of the input signal ( ) can be computed using FFT (Fast Fourier Transform) routines. Convolutions involving individual frequency channels ( ) can be obtained numerically or analytically.

The "normalized mask" of intermodulation noise spectral density can then be determined by normalizing the result relative to the total intermodulation noise power over the frequency band ΔF. The total intermodulation noise power is obtained by integrating the intermodulation noise power spectral density on the ΔF frequency band.

According to another example, the result may be normalized to the aggregated intermodulation noise power over the frequency band ΔF for power values within a predetermined range of values (e.g. for power values between a lower threshold, e.g. -10 dB, and a maximum power value).

Other methods of normalizing the intermodulation noise spectral density could be considered (e.g. normalizing with respect to a maximum power value rather than with respect to an aggregate power value). The choice of a normalization method is only one variation of the invention.

It should be noted that it is also possible to consider intermodulation products of odd order greater than three (for example intermodulation products of order five or seven). These intermodulation products can be calculated by an analytical method similar to that described above for intermodulation products of order three. Taking into account intermodulation products of odd order greater than three makes it possible to obtain better precision for the determination 101 of the normalized intermodulation noise spectral density mask. However, this involves a greater number of calculations and consequently an increase in the time required to execute the method 100 according to the invention.

Instead of precisely calculating intermodulation products of odd order greater than three, it is possible to apply to the intermodulation noise spectral density a correction value to compensate for unaccounted intermodulation products.

This correction value can for example be estimated by simulation by comparing for several examples the spectral density of intermodulation noise calculated by taking into account the intermodulation products of order three and five (or even the intermodulation products of order three, five and seven) with the spectral density of intermodulation noise calculated by taking into account only the intermodulation products of order three. The correction value can then be determined based on an average difference observed between the spectral densities thus calculated.

In the mode of implementation described with reference to the , the normalized intermodulation noise power spectral density mask is the same for all amplifiers. In other words, to calculate the normalized intermodulation noise power spectral density mask of an amplifier, we consider the set of all frequency channels present in the input signals, and we do not take into account the weighting coefficients. This should theoretically not be the case because the frequency channels at the input of the amplifier and the power levels of said frequency channels are not the same from one amplifier to another (because the weighting coefficients associated with the different frequency channels are not the same from one amplifier to another).

Experience shows, however, that this approximation is acceptable because the error introduced into the calculation of the C/IM ratio with this hypothesis remains relatively small. This approximation is all the more valid as the number of sources 24 is large (the power is averaged over all the sources).

It is however also possible not to make this approximation and to take into account, for each amplifier, the weighting coefficients associated with the different frequency channels in the calculation of the normalized intermodulation noise power spectral density mask. This is what is proposed by the embodiment described with reference to the . In this case, a different normalized mask is calculated for each amplifier. This embodiment allows to increase the precision of the calculated C/IM ratio, but it requires a greater number of calculations and therefore a longer execution time.

The following steps of method 100 according to the invention (steps 102 to 110) are identical for the two embodiments described respectively in figures 4 and 5.

The method 100 comprises a step 102 of determining, for each amplifier, an operating point of the amplifier. It is determined from the respective power levels of the frequency channels of the input signals, and from the weighting coefficients respectively associated with said amplifier (i.e. the weighting coefficients associated with the source 24 with which said amplifier is associated). The operating point corresponds to the aggregated power at the output of the amplifier over the entire frequency band ΔF considered.

At the output of step 102, we therefore know the operating point (i.e. the output power) of each amplifier.

The method 100 then comprises a step 103 of determining, for each amplifier, a total intermodulation noise power as a function of the operating point and as a function of the linearity characteristics of the amplifier.

As previously indicated, the linearity characteristics of the amplifier are for example defined by a curve NPR = f(OBO) provided by the amplifier manufacturer. However, it is also possible to obtain the linearity characteristics of the amplifier by simulation methods. This simulation can in particular be implemented by modeling the amplifier with a Shimbo model. The linearity characteristics of the amplifier can then be defined by a curve of type C/IM = f(OBO).

At the output of step 103, the power level with which each amplifier contributes to the intermodulation noise is known, but the spectral distribution of the intermodulation products is not known (the intermodulation products are not uniformly distributed in terms of power). The spectral distribution of the intermodulation products is, however, indicated by the normalized mask determined in step 101.

The method 100 comprises a step 104 of determining, for each amplifier, an “absolute mask” of intermodulation noise power spectral density by weighting the normalized mask (determined in step 101) with the total intermodulation noise power (determined in step 103). An “absolute mask” of intermodulation noise power spectral density is thus determined for each amplifier.

We thus obtain at the output of step 104, for each amplifier, a power spectral density mask of intermodulation noise which is representative of both the spectral distribution and the power levels of the intermodulation products specific to the amplifier.

The method 100 then comprises a step 105 of determining, for each source 24, a “radiated mask” of intermodulation noise power spectral density by weighting the absolute mask (determined in step 104) of the amplifier to which the source is connected by the radiation pattern of the source 24.

Step 105 thus makes it possible to add the spatial dimension, via the radiation diagram. The “radiated mask” thus makes it possible to determine at a given point on the Earth’s surface, for each source 24, a spectral density of intermodulation noise power radiated at this point.

The method 100 then comprises a step 106 of determining a “total radiated mask” of intermodulation noise power spectral density by a power summation of the radiated masks of intermodulation noise power spectral density of the different sources 24. The total radiated mask makes it possible to determine at a given point on the Earth's surface, for all of the sources 24, a radiated intermodulation noise power spectral density at this point.

The use of the sum in power of the radiated masks is based on a simplification which amounts to considering that the intermodulation noises between the different sources are decorrelated, which is theoretically not the case. However, experience shows that this simplification is acceptable because the error introduced in the calculation of the C/IM ratio with this hypothesis remains relatively low in comparison with results provided by theoretically exact models. This simplification makes it possible to reduce the calculation times to provide a modeling of the performances in terms of radiated C/IM in a limited time. It should be noted that the higher the number of intermodulation products, the more the intermodulation noises between the different sources are decorrelated. The method 100 according to the invention is therefore particularly well suited to antennas allowing a large number of beams to be formed simultaneously.

Optionally, a correction value may be applied to the calculated C/IM ratio to take into account the error introduced by the approximation related to the power summation of the radiated masks. This correction value may in particular be determined according to a reuse factor of the frequency channels to form the different desired radio beams. For example, it may be an average reuse rate (corresponding to the average of the reuse rates of the different frequency channels). It has indeed been observed that the higher the average reuse rate, the more it is appropriate to apply a high correction value to the calculated C/IM ratio to compensate for the approximation related to the power summation of the radiated masks. According to another example, the reuse factor may correspond to a difference between the maximum reuse rate and the minimum reuse rate among the different frequency channels. It has in fact been observed that the greater this difference (i.e. the more heterogeneous the reuse rate is for the frequency channels), the more it is appropriate to apply a high correction value to the calculated C/IM ratio to compensate for the approximation linked to the power summation of the radiated masks.

We then consider a particular frequency channel and a particular geographic area of interest (area served by a beam), and we seek to determine the C/IM ratio radiated in this geographic area of interest for the frequency channel considered. For example, we determine the C/IM ratio radiated at a point corresponding to the center of the geographic area. Alternatively, it is possible to determine a C/IM ratio radiated at several points in the geographic area of interest (for example according to a grid of the geographic area of interest).

For this purpose, the method 100 comprises a step 107 of determining a radiated power of intermodulation noise for the geographical area of interest and for the frequency channel considered. This determination step 107 is implemented by an integration, on the frequency sub-band Δf corresponding to said frequency channel, of the total radiated mask of intermodulation noise power spectral density for the geographical area of interest considered.

The method 100 then comprises a step 108 of determining, for the frequency channel considered, an equivalent isotropic power radiated by the active multi-beam antenna 20 (EIRP of the carrier corresponding to the frequency channel). This EIRP can be calculated from the RF power in the carrier and from the radiation patterns and the weighting coefficients.

The method 100 then comprises a determination step 109, for the geographical area of interest and for the frequency channel considered, of a “radiated C/IM” ratio between the radiated equivalent isotropic power (determined in step 108) and the radiated intermodulation noise power (determined in step 107).

Finally, the method 100 comprises a verification step 110 of a predetermined criterion as a function of the “radiated C/IM” ratio thus determined. This step makes it possible to validate the configuration of the payload of the telecommunications satellite 30 in terms of radiated C/IM performance. For example, it is possible to raise an alert (for example via the user interface 14) if the radiated C/IM is below a predefined threshold. Such provisions make it possible to verify whether it is necessary to take measures to modify the configuration of the payload of the telecommunications satellite 30 so that it can perform its mission efficiently. According to another example, it is possible to define a distribution function (CDF, for “Cumulative Distribution Function”) of a radiated C/IM ratio over the entire frequency band ΔF considered and for the entire coverage region considered. For example, for a point in the coverage region, it is possible to determine by which carrier it is illuminated, and to calculate the radiated C/IM ratio for the carrier in question at the point considered. If the point considered is illuminated by several carriers, it is possible to calculate an average value, or a value corresponding to a worst case, of the radiated C/IM ratio at the point considered. Such a repair function makes it possible, for example, to determine what percentage of the coverage region has a radiated C/IM ratio greater than or equal to a predefined threshold.

The above description clearly illustrates that, through its various features and their advantages, the present invention achieves the set objectives. In particular, the proposed solution makes it possible to calculate a radiated C/IM ratio in a limited time in order in particular to allow its use during operational phases, while ensuring a sufficient level of precision. The complexity of evaluating the C/IM ratio is linked to the high number of carriers and beams required to ensure the mission of the telecommunications satellite 30. The proposed method is therefore particularly well suited to evaluating the C/IM ratio in cases of complex payload (the other calculation methods available in the prior art then being too long to implement). The proposed method uses a global and analytical approach which makes it possible to avoid numerous unit calculations and thus maintain an almost negligible execution time.

Claims (12)

Method (100) implemented by a computer (10) for validating a configuration of a payload of a telecommunications satellite (30), said satellite (30) being configured to transmit a signal with an active multi-beam antenna (20) comprising a beamforming network (25), the beamforming network (25) being connected to a plurality of sources (24) and adapted to simultaneously form several beams of interest serving geographical areas on the surface of the Earth, the beamforming network (25) being configured to receive as input, for each beam of interest to be formed, an input signal comprising one or more frequency channels with respective power levels, and to transmit to each source (24) a linear combination of the input signals weighted by weighting coefficients, each source (24) being connected to an amplifier,
method (100) comprises for each amplifier:

• a determination (101) of a “normalized mask” of intermodulation noise power spectral density from the frequency channels composing the input signals and the respective power levels of said frequency channels,
• a determination (102) of an operating point corresponding to an output power of the amplifier, the operating point being determined as a function of the power levels of the frequency channels of the input signals, and as a function of the weighting coefficients associated with said amplifier,
• a determination (103) of a total intermodulation noise power as a function of the operating point and linearity characteristics of the amplifier,
• a determination (104) of an “absolute mask” of intermodulation noise power spectral density by weighting the normalized mask with the total intermodulation noise power,

method (100) includes for each source (24):

• a determination (105) of a "radiated mask" of intermodulation noise power spectral density from the absolute mask of intermodulation noise power spectral density of the amplifier to which the source (24) is connected, and from a radiation pattern of the source,

method (100) further comprises:

• a determination (106) of a “total radiated mask” of intermodulation noise power spectral density by a power summation of the radiated masks of the different sources (24),

for at least one frequency channel of interest, and for at least one geographic area of interest:

• a determination (107) of a radiated power of intermodulation noise by an integration, over a frequency sub-band corresponding to said frequency channel, of the total radiated mask of intermodulation noise power spectral density for the geographical area of interest,
• a determination (108), for the frequency channel of interest, of an equivalent isotropic power radiated by the active multi-beam antenna (20),
• a determination (109) of a “radiated C/IM” ratio between the radiated equivalent isotropic power and the radiated intermodulation noise power,
• a verification (110) of a predetermined criterion as a function of the “radiated C/IM” ratio thus determined.
  

The method (100) of claim 1 wherein the normalized intermodulation noise power spectral density mask is common to all amplifiers. The method (100) of claim 1 wherein the normalized intermodulation noise power spectral density mask is determined for an amplifier based on weighting coefficients associated with said amplifier. Method (100) according to any one of claims 1 to 3 wherein the determination (101) of the normalized intermodulation noise power spectral density mask comprises a calculation of third order intermodulation products generated by the frequency channels of the input signals. Method (100) according to claim 4 in which the determination (101) of the normalized intermodulation noise power spectral density mask further comprises a calculation of intermodulation products of odd order strictly greater than three generated by the frequency channels of the input signals. Method (100) according to claim 4 in which the determination (101) of the normalized intermodulation noise power spectral density mask comprises an application of a correction value aimed at compensating for intermodulation products of order strictly greater than three not taken into account. Method (100) according to any one of claims 1 to 6 in which the determination (101) of the normalized intermodulation noise power spectral density mask comprises a normalization calculation as a function of the result of an integration, over a frequency band, of the intermodulation noise power spectral density. Method (100) according to any one of claims 1 to 6 in which the determination (101) of the normalized intermodulation noise power spectral density mask comprises a normalization calculation as a function of the result of an integration, over a frequency band, of the intermodulation noise power spectral density for power values included in a predetermined range of values. Method (100) according to any one of claims 1 to 8 comprising an application of a correction value to the “radiated C/IM” ratio, the correction value being determined as a function of a reuse factor of the frequency channels to form the beams of interest. Method (100) according to any one of claims 8 or 9 wherein the frequency band is one of the bands W, Q, V, Ka, Ku, L, S, C. Computer program product (13) comprising instructions which, when executed by a computer (10), cause the latter to implement the steps of the method (100) according to any one of claims 1 to 10. A computer-readable recording medium (12) comprising instructions which, when executed by a computer, cause the computer to carry out the steps of the method (100) according to any one of claims 1 to 10.