RU2832598C1 - Method of processing signal in space-based ku-band scatterometer with rotating "fan" beam pattern - Google Patents

Abstract

FIELD: radio engineering.

SUBSTANCE: invention relates to methods of processing signals in radar remote sensing of the earth's surface using space-based scatterometers. In the disclosed method of processing signals in the on-board equipment of the scatterometer, the specific effective scattering area of the global ocean surface is measured at different azimuthal angles by means of emission, reception and correlation processing of probing pulses of microwave range with linear frequency modulation, anticipatory compensation of average Doppler frequency shift during pulse emission. Frequency deviation of the chirp signal is changed depending on the angle of rotation of the antenna at constant signal duration according to the law ΔF=ΔF₀+ƒ_d·k, where ΔF₀ is the average frequency deviation, ƒ_d is the expected average Doppler frequency shift, k is a constant which depends on the orbital parameters of the spacecraft motion and the geometric parameters of the irradiation spot as follows.

EFFECT: reduced measurement error of specific effective scattering area under condition of radiated power limitation due to preservation of geometric parameters and orientation of resolution elements during rotation of the beam pattern.

1 cl, 4 dwg

Description

The invention relates to the field of radio engineering, and more specifically to methods of processing signals in radar remote sensing of the earth's surface using space-based scatterometers. Space-based scatterometers provide operational global measurement of the specific effective scattering area (SESA) of sections of the world ocean surface at various azimuthal angles using the emission, reception and processing of microwave probing pulses. The information obtained is used primarily to determine the parameters of the surface (near-surface) wind.

To ensure an acceptable error in measuring the ERCS taking into account the statistical features of microwave radiation scattering by the sea surface within the irradiation spot, it is necessary to accumulate and incoherently average the results of ERCS measurements of a large number of small surface areas. For this purpose, in space-based scatterometers, the received signal is divided by correlation processing into signals reflected from the resolution elements within the irradiation spot. Resolution elements will be understood as surface areas on the boundary of which the level of the normalized autocorrelation function (ACF) is higher than a certain threshold, usually minus 3 dB.

It is known [1 - Lecompte P. The ERS Scatterometer Instrument and the On-Ground Processing of its Data // Proceedings of the ESA Workshop on Emerging Scatterometer Applications, Noordwijk, The Netherlands, 5-7 October, 1998, pp. 241-260] that the error in measuring the ERS is characterized by a normalized root-mean-square deviation

,

where σ is the value of the UEPR;

- standard deviation of the UEPR estimate;

N - number of independent measurements;

σ _шэ - noise equivalent of the UEPR.

The noise equivalent is such a value of the NRCS, at which the energy of the received and subjected to correlation processing signal from the surface area is equal to the spectral density of the noise power. As the signal level decreases, the noise equivalent of the NRCS increases. The number of independent measurements N corresponds to the number of non-overlapping resolution elements.

The power consumed by a space-based scatterometer is strictly limited and, consequently, the emitted power is also limited. Therefore, a decrease in the resolution element size leads, on the one hand, to an increase in the number of independent measurements N, and, on the other hand, to an increase in σ _шэ .

When developing a scatterometer, the optimal resolution element sizes are selected to ensure the minimum integral value of Kp in the required range of variation of the measured ERCS, with a limitation on the power consumed by the scatterometer. Deviation of the resolution element sizes from the optimal ones usually leads to deterioration in the quality of ERCS measurements. The duration of the probing pulse is selected as long as possible in order to reduce the peak radiated power.

A method for processing the received signal in a space-based scatterometer with a rotating “fan” directivity pattern (DP) of the C-band is known [2 - United States Patent 6,137,437. Lin C.-C, Wilson J.J.W., Impagnatiello F., Park P. Spaceborne scatterometer. October 24, 2000].

However, this invention does not take into account the essential features of processing a signal reflected from a section of great length in the radial direction, namely: a significant difference in the Doppler frequency shifts for different sections of the irradiation spot and different azimuthal angles of rotation of the antenna.

As is known, the multiplication of the received chirp signal by the reference chirp signal and the subsequent Fourier analysis are collectively equivalent to the correlation processing procedure [3 - Skolnik M. Radar handbook. - N.Y.: McGraw-Hill, 2008 - pp. 8.31-8.36.]. In this case, the uncertainty region in the frequency-time plane is determined by a combination of the frequency deviation and the duration of the probing pulse. At the same time, when the beam pattern rotates, the distribution of the Doppler frequency shift within the irradiation spot changes significantly. Thus, if the pulse duration and the rate of frequency change remain constant, the frequency-time parameters of the probing and reference pulses do not change. This leads to the fact that the geometric parameters of the resolution element, determined by the frequency-time parameters of the reflected signal, will change significantly.

The consequence of changing the geometric parameters of the resolution element depending on the position of the resolution element within the irradiation spot and on the rotation angle of the DN is an increase in the error in measuring the UEPR. In this regard, the specified scatterometer has limited industrial applicability.

A method for signal processing for resolving surface areas and taking into account the influence of the Doppler frequency shift is known, adopted as a prototype [4 - Patent CN 101363913 B. Signal processing method of fan-shaped beam circular cone scanning microwave scatterometer, 27.09.2008]. This method is used in a space-based Ku-band scatterometer with a rotating "fan" directional pattern [5 - Lin W. Dong X. Design and Optimization of Ku-band Rotating, Range-Gated, Fanbeam Scatterometer // International Journal of Remote Sensing, Vol. 32, No. 8, July 2011, pp. 2151-2171].

The method is based on the following operations:

- linear frequency modulation of the probing pulse, with constant duration and rate of frequency change;

- anticipatory compensation of the average Doppler frequency shift during pulse emission;

- determining the boundary frequencies corresponding to the edges of the resolution elements, taking into account the expected reception time and the Doppler frequency shift, in the intermediate signal after multiplying the received signal by the reference chirp signal;

- summation of the energy of the Fourier components of the intermediate signal in frequency intervals corresponding to the resolution elements.

However, as already indicated above, each Fourier component is the result of performing the operation of calculating the correlation integral. Summation of the energy of the Fourier components corresponds to incoherent averaging of the results of signal energy measurements.

From the point of view of using this method of signal processing in a space-based scatterometer, when the “fan” DN rotates, there are the following disadvantages:

- only the preservation of the positions of the centers is ensured, but not the preservation of the sizes and orientation of the resolution elements;

- the number of independent measurements N changes when the antenna is rotated due to a change in the size of the resolution elements themselves.

As a consequence, the number of independent measurements N does not correspond to the optimal one, which in turn leads to an increase in the measurement error of the UEPR. Analysis of the radiometric accuracy of the scatterometer using the proposed signal processing method shows that the parameter Kp does change significantly depending on the rotation angle [6 - Wenming Lin, Xiaolong Dong, Di Zhu. Backscatter measurement accuracy of a spaceborne rotating fan-beam scatterometer // EUMETSAT/ESA Scatterometer Science Conference, Darmstadt, Germany, 11-13 April 2011]. Thus, the noted processing method, selected as a prototype, also does not take into account all the features of processing the signal reflected from a large area in the radial direction, and thus does not ensure the preservation of the size and orientation of the resolution element and, as a consequence, leads to an increase in the measurement error.

The task, which the claimed invention is aimed at solving, is to reduce the error in measuring the UEPR using a space-based Ki-band scatterometer with a rotating “fan” radiation pattern, subject to limitations on the radiated power.

To solve the specified problem, a method is proposed for processing a signal in a space-based Ku-band scatterometer with a rotating “fan” directional pattern, in which the specific effective scattering area of the world ocean surface is measured at various azimuthal angles using the emission, reception and correlation processing of microwave probing pulses with linear frequency modulation, with advanced compensation for the average Doppler frequency shift during pulse emission.

According to the invention, when processing a signal in the on-board equipment of a scatterometer, depending on the angle of rotation of the antenna with a constant signal duration, the frequency deviation of the chirp signal is changed according to the law

ΔF=ΔF ₀ +ƒ _d ⋅k,

where ΔF ₀ is the average frequency deviation;

ƒ _d - expected average Doppler frequency shift;

ƒ _d =ƒ ₁ +ƒ ₂ ⋅cos(ϕ _A6a -ϕ _A0 )+ƒ ₃ ⋅cosϕ _ka ⋅sin(ϕ _A6a -ϕ _A0 ),

where ƒ _1, ƒ ₂ , ƒ ₃ are constants with the dimension of frequency;

ϕ _Aba is the angle of rotation of the antenna according to the measurements of the sensors relative to the abscissa axis of the scatterometer coordinate system (CS);

ϕ _A0 is the angle between the direction of the spacecraft movement in the scatterometer coordinate system and the abscissa axis of this coordinate system;

ϕ _КА - the angle of elevation of the КА, measured from the plane of the equator.

k is a constant, k≥0, which depends on the orbital parameters of the spacecraft's motion and the geometric parameters of the irradiation spot as follows

where τ is the duration of the probing pulse;

c - speed of light;

β - angle of incidence;

L - slant range to the resolution element;

R ₃ is the radius of the Earth.

The technical result is a reduction in the measurement error of the specific effective scattering area under the condition of limiting the radiated power due to maintaining the geometric parameters and orientation of the resolution elements during rotation of the beam pattern.

The combination of distinctive features and properties of the proposed method is unknown from the current level of technology, therefore it meets the criteria of “novelty” and “inventive step”.

Fig. 1 shows a diagram of the surface probing by a space-based scatterometer.

Fig. 2 shows an example of the distribution of azimuthal observation angles of scattering cells.

Fig. 3 shows the arrangement of the resolution elements within the irradiation spot.

Fig. 4 shows a structural diagram of a device implementing the claimed method.

Let us give a more detailed description of the signal processing method in a space-based Ku-band scatterometer with a rotating "fan" directional pattern. As in the prototype [4], the rotation of the "fan" directional pattern allows obtaining measurements of the sea surface ERCS at different azimuthal angles using one antenna system while ensuring the continuity of the survey band. The probing scheme is shown in Fig. 1. As in the prototype, complex signals are used to improve the resolution, in particular the chirp signal. The received signal is multiplied by the reference signal, which allows measuring the ERCS value.

However, unlike the prototype, with given parameters of the “fan” directional pattern, the frequency deviation of the emitted signal changes with its constant duration depending on the angle of rotation of the antenna.

When implementing the proposed method, the following sequence of operations is performed:

- using a rotating "fan" antenna pattern, the width of which in the elevation and azimuthal planes, the angle of inclination of the electric axis of the antenna are determined by the orbit parameters and the required number of independent observations of the resolution elements at different angles, they produce radiation, reception of chirp signals reflected from the sea surface and their subsequent correlation processing;

- when rotating the beam, the frequency deviation of the emitted chirp signals changes according to the law:

ΔF=ΔF ₀ +ƒ _d ⋅k,

where ΔF ₀ is the average frequency deviation;

ƒ _d - expected average Doppler frequency shift;

k is a constant, k≥0, which depends on the orbital parameters of the spacecraft (SC) motion and the geometric parameters of the irradiation spot.

The comparative analysis of this method and the prototype shows that the claimed method additionally introduces an operation for changing the frequency deviation of the chirp signal depending on the antenna rotation angle, the value of which is determined by the value of the expected Doppler frequency shift depending on the orbit parameters.

Let us consider in more detail the essence of the claimed method. As is known, in the case of using the results of the NRCS measurements to reconstruct the parameters of the surface wind, the optimal angles of incidence should be considered to be angles from 30 to 60 degrees [7 - Wentz F., Smith D. A model function for the ocean-normalized radar cross section at 14 GHz derived from NSCAT observations // Journal of Geophysical Research, Vol. 104, No. C5, pp. 11,499-11,514. May 15, 1999]. These are the angles at which the dependence of the NRCS on the angle between the direction of sounding and the direction of the surface wind is best manifested. The altitude of the sun-synchronous orbit H is usually specified by the developer of the spacecraft and is from 600 to 900 km. The distance from the subsatellite point to the outer edge of the irradiation spot R ₂ (Fig. 1) corresponds to the edge of the survey strip. Therefore, the value of R ₂ is chosen to be maximum, but such that the angle of incidence does not exceed 60 degrees. At the same time, for angles greater than 50 degrees, the value of the UEPR for horizontal polarization rapidly decreases with increasing angle of incidence at low values of the surface wind speed. From the point of view of reconstructing weak winds, taking into account the above-mentioned requirements for choosing the sounding angle and using two polarizations for the angle θ ₂ , which determines the outer boundary of the irradiation spot (Fig. 1), a value close to 50 degrees is chosen.

The distance from the sub-satellite point to the inner edge of the irradiation spot R ₁ is selected from the following considerations. The distribution of azimuthal observation angles ϕ is determined by the condition

,

where γ is the offset from the center of the viewing strip in the transverse direction from the path line;

R is the distance from the sub-satellite point to the observation point.

For a system with a rotating "fan" DN

R ₁ ≤R≤R ₂ ,

-R ₂ ≤γ≤R ₂ .

Therefore, all points for observation angles ϕ lie between two curves:

,

for γ≤R ₁ and

,

for R ₁ ≤γ≤R ₂ .

The greater the value of R ₁ , the fewer observations can be obtained at different angles. On the other hand, the width of the antenna pattern decreases, the antenna gain increases, and, consequently, with the same radiated power, the level of the reflected signal increases and the noise equivalent of the NRCS decreases. The optimal range of variation of the parameter R ₁ lies within the limits 0.75R ₂ ≤R ₁ ≤0.85R ₂ .

To refine the value of R ₁ , we take into account that in the scatterometer [8 - Figa-Saldaña J., Wilson JJW, Attema E., Gelsthorpe R., Drinkwater MR, Stoffelen A. The advanced scatterometer (ASCAT) on the meteorological operational (MetOp) platform: A follow-up for European wind scatterometers // Can. J. Remote Sensing - V. 28. - N.3. - 2002. - PP. 404-412.] the area in which the effective retrieval of the surface wind parameters is carried out lies in two bands, each 550 km long. In the scatterometer [9 - Spencer MW, Wu C, Long DG Improved Resolution Backscatter Measurements with the SeaWinds Pencil-Beam Scatterometer // IEEE Transactions On Geoscience And Remote Sensing - V. 38. - N.1. - January 2000. - PP. 89-104.] The swath is 1800 km, but effective wind recovery can only be achieved within 1150 km. Thus, R ₁ can be chosen to be 0.85R ₂ .

The required rotation period of the DN is determined from the condition:

where m=4 is the required minimum number of observations at different azimuthal angles;

ν _A is the speed of movement of the sub-satellite point on the Earth's surface.

An example of the distribution of azimuthal angles of observation of scattering cells for R ₁ = 730 km, R ₂ = 900 km is shown in Fig. 2. In this case, the rotation period is _TA = 10.8 s. The size of the swath element in which the wind speed and direction are determined is chosen to be 25 km [10 - Satellites oceanography and society / edited by D. Halpern - Elsivier, 2000, 368 p.].

The initial approximation for the direction of the maximum and the width of the RP in the elevation plane is chosen so that for the angles θ ₁ and θ ₂ of sighting the outer and inner edges of the irradiation spot from the nadir, the level of the antenna gain relative to the maximum is approximately equal to minus 2.2 dB. Specific values of the direction of the maximum and the width of the RP are obtained based on the optimization results, based on the criterion of the maximum of the minimum power of the received signal, taking into account the change in range and energy losses due to the rotation of the RP during the signal propagation time.

The beam pattern width in the azimuth plane is chosen to be as small as possible. It is limited by both the maximum permissible size of the antenna system and the permissible level of power loss due to the beam pattern rotation during signal propagation.

The duration of the probing pulse is limited from above by the time of signal propagation to the inner edge of the irradiation spot. It is chosen to be maximum to increase the energy signal/noise ratio in the received signal. The repetition period of the probing pulses is limited by the time of signal propagation to the outer edge of the irradiation spot. It is chosen to be minimum to increase the number of independent measurements in each element of the viewing band.

To ensure the necessary separation of the signal into the corresponding signals reflected from different resolution elements, a chirp signal is used. As in the prototype, the use of a chirp signal is due to the possibility of full digital signal processing on board the spacecraft with low requirements for computing resources.

If we neglect the change in the slope of the chirp signal due to the Doppler effect, and take into account only the frequency shift of the signal, then the condition of equality of the ACF to a certain selected level, which determines the boundary of the resolution element, can be written as

,

where τ is the duration of the probing pulse;

Δƒ _d (r, ϕ) - Doppler frequency shift;

Δt(r, ϕ - signal propagation delay;

ΔF - bandwidth of the probing pulse;

r is the position of the center of the resolution element, measured from the sub-satellite point along the Earth's surface in the elevation direction;

p - azimuthal position of the center of the resolution element.

With the length of the resolution element along the elevation direction equal to , the location of the resolution elements within the irradiation spot for the case of H=650 km, _R2 =900 km and DN dimensions of 1°×6° is shown in Fig. 3. In this case, the optimal value , corresponding to the ACF level of minus 3 dB, is 0.31 km. However, for different resolution elements, the value of Δϕ _d (r, ϕ) is different. Let us determine the relationship between the change in the value of the Doppler frequency shift and the change in the position of the resolution element.

The condition that determines the displacement of the resolution element along the Earth's surface along the elevation direction by half its length has the following form:

With a linear approximation of the form

you can get that

where c is the speed of light;

β - angle of incidence;

L - slant range to the resolution element;

R ₃ is the radius of the Earth.

In more general terms, we can write that

where k ₁ and k ₂ are constant coefficients.

It follows that when the Doppler frequency shift changes, the length of the resolution element changes. This circumstance was not properly taken into account in the prototype. Since the width of the resolution element is limited by the width of the irradiation spot in the azimuthal plane and does not change when the beam pattern rotates, then to ensure a minimum error in measuring the ERCS by maintaining the dimensions of the resolution elements, it is necessary to change the frequency deviation of the probing pulse according to the law:

ΔF=ΔF ₀ +ƒ _d ⋅k,

where ΔF ₀ is the average frequency deviation;

ƒ _d - expected average Doppler frequency shift;

k is a constant that depends on the orbital parameters of the spacecraft motion and the geometric parameters of the irradiation spot, k≥0,

.

The Doppler frequency shift depending on the antenna rotation angle, taking into account the sun-synchronous orbit, is calculated using the formula:

ƒ _d =ƒ ₁ +ƒ ₂ ⋅cos(ϕ _A6a -ϕ _A0 )+ƒ ₃ ⋅cosϕ _ka ⋅sin(ϕ _A6a -ϕ _A0 ),

where ƒ ₁ , ƒ ₂ , ƒ ₃ are constants with the dimension of frequency;

ϕ _Aba is the angle of rotation of the antenna according to the measurements of the sensors relative to the abscissa axis of the scatterometer;

ϕ _A0 is the angle between the direction of the spacecraft movement in the scatterometer coordinate system and the abscissa axis of this coordinate system;

ϕ _ка - the angle of elevation of the spacecraft, measured from the equatorial plane.

ϕ _A0 is determined by the quaternions of the spacecraft orientation in the orbital coordinate system and the scatterometer orientation in the spacecraft coordinate system. The spacecraft elevation angle is determined by the spacecraft coordinates in the PZ-90 geocentric coordinate system.

Thus, all operations to determine the required frequency parameters of the probing signal can be performed on board the spacecraft using the onboard scatterometer equipment.

The expected average Doppler frequency shift is used, as in the prototype, for the preemptive compensation of the Doppler shift in the probing signal. In this case, the signal in the receiving channel does not shift in frequency.

The reference signal is selected with the same linear frequency modulation rate ΔF/τ as the emitted probing pulse.

To implement all operations, the following structural diagram of the scatterometer is proposed, shown in Fig. 4.

In the digital signal processing unit (DSPU) 1, depending on the value of the antenna rotation angle received from the rotation drive (RD) 2, a probing chirp signal is formed. After the probing pulse is formed, the signal frequency is converted upward in the mixer (SM1) 3 and then amplified in the power amplifiers (PA) 4 and transmitted via the waveguide path (WP) 5 to the antenna system (AS) 6. When the reflected signal is received, it is amplified in the low-noise power amplifier (LNA) 7, the signal frequency is converted downward in the mixer (SM2) 8 and then digitized and correlated in the DSPU 1. The harmonic signal of the local oscillator of the reference oscillator (RO) 9 is used to convert the signal in the mixers (SM1) 3 and (SM2) 8 and to form a clock signal in the DSPU 1.

The technical result is achieved by selecting the optimal size of the DP; ensuring optimal geometric parameters of the resolution elements that are preserved during the rotation of the DP due to:

- forward compensation of the average Doppler frequency shift ƒ _d ;

- changes in the frequency deviation of the probing chirp signal according to the law

ΔF=ΔF ₀ +ƒ _d ⋅k

where ΔF ₀ is the average frequency deviation;

k is a constant that depends on the orbital parameters of the spacecraft’s motion and the geometric parameters of the irradiation spot;

- ensuring the appropriate rate of linear frequency modulation ΔF/τ in the reference signal during correlation processing.

Thus, as a result of introducing an additional operation of changing the frequency deviation of the chirp signal depending on the angle of rotation of the antenna, the value of which is determined by the value of the expected Doppler frequency shift depending on the orbit parameters, the following technical result is achieved: the error in measuring the specific effective scattering area is reduced under the condition of limiting the radiated power due to the preservation of the geometric parameters and orientation of the resolution elements during rotation of the radiation pattern.

Claims (11)

A method for processing a signal in a space-based Ku-band scatterometer with a rotating "fan" radiation pattern, in which the specific effective scattering area of the world ocean surface is measured at various azimuthal angles using the emission, reception and correlation processing of microwave probing pulses with linear frequency modulation, advanced compensation for the average Doppler frequency shift during pulse emission, characterized in that the frequency deviation of the probing signal is changed depending on the azimuthal rotation of the antenna according to the law ΔF=ΔF ₀ +ƒ _d ⋅k where ΔF ₀ is the average frequency deviation; ƒ _d - expected average Doppler frequency shift; k is a constant, k≥0, which depends on the orbital parameters of the spacecraft's motion and the geometric parameters of the irradiation spot as follows where τ is the duration of the probing pulse; c - speed of light; β - angle of incidence; L - slant range to the resolution element; R _З is the radius of the Earth.

Images