CN113703008A - Method for improving sea surface survey high precision based on coherent integration time optimization model - Google Patents

Abstract

The invention discloses a method for improving the high precision of sea surface measurement based on a coherent integration time optimization model. ; According to the prediction and measurement high-precision model, the coherent integration time optimization model is constructed; according to the coherent integration time optimization model, the variation curves of the prediction and measurement accuracy with the coherent integration time under different scenarios are calculated respectively; For the minimum value point, the estimated measurement accuracy corresponding to the minimum value point is used as the output of the optimal measurement accuracy. The invention aims to improve the measurement accuracy of iGNSS-R, constructs the coherent integration time optimization model by deriving the conversion relationship between the coherent integration time, the waveform correlation and the measurement accuracy, and then applies the coherent integration time optimization model The variation of the accuracy with the coherent integration time is more accurately estimated to optimize the accuracy of the final sea surface altimetry result.

Description

A method for improving sea surface measurement accuracy based on coherent integration time optimization model

technical field

The invention belongs to the cross technical fields of satellite altimetry, oceanographic mapping and the like, and particularly relates to a method for improving the high precision of sea surface measurement based on a coherent integration time optimization model.

Background technique

As one of the important parameters for marine ecosystem monitoring, accurate measurement of sea level changes is of great significance to applications such as fishery, oil drilling, and commercial navigation. Compared with traditional tide gauge stations and satellite radar altimetry, the passive reflection measurement and interference system (PARIS) altimetry method proposed in 1993 has unique advantages: (1) The receiver can simultaneously capture signals from multiple GNSS satellites, which greatly improves the accuracy of altimetry. (2) As a new dual-base passive remote sensing method, it has the advantages of low cost, low power consumption, all-weather and high time revisit rate, which can make up for the shortcomings of the existing marine remote sensing technology to a large extent. Since the technology was proposed, its effectiveness has been verified through shore-based, airborne, and spaceborne altimetry experiments.

There are also a series of problems in the existing satellite altimetry technology: (1) In terms of high accuracy, the traditional satellite antenna only considers the receiving gain of the direct signal, resulting in a low signal-to-noise ratio of the received sea surface reflected signal. And compared with the dedicated single-station radar altimeter, the transmission bandwidth of the civil modulation code is narrower, resulting in a smaller slope of the leading edge of the peak energy waveform. (2) In terms of along-orbit spatial resolution, taking GPS signals as an example, the traditional GNSS-R altimetry method uses the C/A code phase delay of the GPS reflected signal relative to the direct signal to invert the sea surface height (CGNSS). In this method, the intersection area of the first equal delay line and the first equal Doppler line is used as the main observation area for altimetry, and the footprint corresponding to this area is about 10 km in size. In order to further reduce the influence of thermal noise and speckle noise on the measurement accuracy, coherent integration in milliseconds and incoherent accumulation in seconds are often used to process signals to improve the signal-to-noise ratio. Considering the velocity of the satellite's sub-satellite point (km/s), the actual measured single sea surface height will represent the height average over a larger area, which further reduces the spatial resolution along the orbit. Studies have shown that in the spaceborne altimetry scenario, in order to achieve a measurement accuracy better than 20cm, the signal processing time is about 10s, and the spatial resolution along the orbit is about 65km. The realization of sea surface height observation with high precision and high along-orbit spatial resolution will provide important data and information resources for earth science research such as monitoring of small and medium-scale ocean phenomena, establishment of ocean gravity field models with high temporal and spatial resolution, and global or regional ocean tide models.

The main observation of GNSS-R is the time-delay Doppler map (DDM) after the reflected signal is processed. Fernando detailed the corresponding relationship between the DDM map and its spatial position. Satellite altimetry usually only uses the waveform at 0 Doppler, and the time delay information of the reflected signal needs to be inverted from this waveform. Commonly used inversion algorithms include DER, MAX, and HALF. The DER algorithm defines that the mirror point delay is located at the maximum point of the derivative of the waveform, and Rius gives the specific derivation process of the algorithm; the MAX algorithm defines that the mirror point delay is located at the peak point of the waveform, and the algorithm will bring about a rough sea surface. Larger error; the HALF method takes the point corresponding to 75% of the peak correlation power on the waveform as the specular reflection point, and the algorithm is derived from the traditional monostatic radar technology. Mashburn considers and corrects the error terms in the retrieved delay information including transmitter and receiver orbits, ionospheric and tropospheric delay models, and zenith-to-nadir antenna baseline offsets. For the evaluation of altimetry results, Li considers two aspects of high precision and altimetry accuracy. The height measurement accuracy is caused by systematic errors, and the measurement accuracy is mainly caused by the randomness of the received signal caused by thermal noise and speckle noise. Theoretically speaking, the higher the signal-to-noise ratio of the signal, the more accurate the extracted specular point delay information, and the better the measurement accuracy.

In order to improve the measurement accuracy, optimization can be carried out from two aspects of load and signal post-processing. The former method to improve the accuracy includes increasing the gain of the antenna, optimizing the antenna pointing and receiver bandwidth, etc. For the post-processing of the signal, the method to improve the signal-to-noise ratio is to increase the coherent integration time of the signal and the number of incoherent accumulations, both of which have a weakening effect on the noise introduced by different channels. The former is to suppress the thermal noise introduced by the receiver, and the latter is to suppress the speckle noise introduced in the blaze area near the specular reflection point. The total integration time of signal processing is the product of coherent integration time and incoherent accumulation times. In theory, the longer the integration time, the higher the signal-to-noise ratio and the higher the measurement accuracy. However, the inverse relationship between the along-orbit resolution of altimetry and the integration time makes the latter unable to increase infinitely. For altimetry tasks that require high along-orbit resolution, it is usually necessary to sacrifice a certain degree of accuracy. To obtain the highest measurement accuracy under a given spatial resolution requirement, the key is to optimize the combination of coherent integration and incoherent accumulation parameters. Accordingly, Martin-Neira et al. constructed a measurement accuracy conversion formula with coherent integration time as a variable, but they did not consider the influence of the correlation between waveforms. You et al. constructed a waveform correlation model from the time domain and frequency domain respectively to predict the upper limit of the waveform coherence time. The measurement accuracy model constructed by Li et al. considering waveform correlation is in good agreement with the actual measurement results, but they did not discuss the influence of coherent integration time on measurement accuracy in detail. At this stage, the processing of intermediate frequency data often adopts empirical parameters (the coherent integration time is 10s in the airborne case, and the coherent integration time is 1ms in the spaceborne scenario).

SUMMARY OF THE INVENTION

The technical solution of the present invention is to overcome the deficiencies of the prior art and provide a method for improving the high precision of sea surface measurement based on a coherent integration time optimization model, aiming at improving the high precision of iGNSS-R measurement. According to the conversion relationship between the performance and the measurement accuracy, the coherent integration time optimization model is constructed, and then the coherent integration time optimization model is used to more accurately estimate the variation law of the accuracy with the coherent integration time, so as to optimize the final sea surface altimetry. The precision of the result.

In order to solve the above technical problems, the present invention discloses a method for improving the high precision of sea surface measurement based on a coherent integration time optimization model, including:

According to the transformation principle of the measurement high-precision formula, the uncertainty of the energy waveform at the mirror point is converted to obtain the pre-measured high-precision model;

According to the pre-measured high-precision model, the coherent integration time optimization model is constructed and obtained;

According to the coherent integration time optimization model, the variation curves of the pre-measured high precision with the coherent integration time under different scenarios are calculated respectively;

The minimum value points are obtained from the change curve, and the estimated measurement accuracy corresponding to the minimum value point is used as the optimal measurement accuracy output.

In the above-mentioned method for improving the high precision of sea surface measurement based on the coherent integration time optimization model, the expression of the obtained high precision prediction model is as follows:

为Z(τ)的平均值，

为

的导数。Among them, σ _h (τ) represents the pre-measurement accuracy, τ represents the time delay, c represents the speed of light in vacuum, i represents the incident angle, σ _Z (τ) represents the mean square error of the incoherent average energy waveform, _Sh (τ ) represents the altimetry sensitivity, Z(τ) represents the incoherent average energy waveform,

is the mean value of Z(τ),

for

derivative of .

In the above method for improving the high accuracy of sea surface measurement based on the coherent integration time optimization model, the calculation formula of the height measurement sensitivity _Sh (τ) is as follows:

和

In the above method for improving the accuracy of sea surface measurement based on the coherent integration time optimization model, the following methods are used to determine

and

The direct signal and the reflected signal are represented by the signal component and the noise component, and the coherent integration process is performed to obtain the complex power waveform y(nT _c ,τ) of the nth coherent integration:

y(nT _c ,τ)=y _s (nT _c ,τ)+y _nd (nT _c ,τ)+y _nr (nT _c ,τ)+y _ndr (nT _c ,τ)…(2)

Among them, n represents the coherent integration times, T _c represents the coherent integration time, nT _c represents the process of the nth coherent integration, y _s (nT _c , τ) represents the cross-correlation energy value of the useful signal term, y _nd (nT _c , τ) represents the cross-correlation energy value of the noise term of the direct signal and the reflected signal, y _nr (nT _c ,τ) represents the cross-correlation energy value of the noise term of the direct signal and the reflected signal, and y _ndr (nT _c ,τ) represents the noise of the direct signal The cross-correlation energy value of the term and the reflected signal noise term;

According to the obtained y(nT _c ,τ), determine the single energy waveform z(nT _c ,τ) of the nth coherent integration:

z(nT _c ,τ)=y(nT _c ,τ)y ^* (nT _c ,τ)…(3)

Among them, y ^* (nT _c ,τ) represents the conjugate of y(nT _c ,τ);

According to z(nT _c ,τ), the incoherent average energy waveform Z(τ) is obtained by solving:

Among them, _NI represents the incoherent average, and <> represents the average value;

Simultaneous equations (2), (3) and (4) yield Z(τ) in terms of coherent integration time:

Z(τ)=<|y _s (nT _c ,τ)| ² >+<|y _nr (nT _c ,τ)| ² >+<|y _nd (nT _c ,τ)| ² >+<|y _ndr (nT _c ,τ)| ² >…(5)

对

求导，得到

According to formula (5), average Z(τ) to get

right

seek guidance, get

In the above method for improving the accuracy of sea surface measurement based on the coherent integration time optimization model, y _s (nT _c ,τ), y _nd (nT _c ,τ), y _nr (nT _c ,τ) and y _ndr (nT _c ,τ) The coherent integration time expression for τ) is as follows:

表示散射点向量，p表示散射点面积，σ₀表示散射截面，Δτ表示散射点时延

与τ的差值，tri()表示三角函数，

表示天线增益，

表示发射机到散射点的距离，

表示接收机到散射点的距离，Δf表示散射点的多普勒频率

与0多普勒的差值，k表示采样次数，B表示接收机的等效噪声带宽，T_{rec_r}表示下视链路的等效输入噪声温度，T_{rec_d}表示上视链路的等效输入噪声温度。Among them, P _d is the total energy of the direct signal at the input of the correlator, P _t is the energy of the transmitted signal, P _r is the total energy of the reflected signal at the input of the correlator,

is the scattering point vector, p is the area of the scattering point, σ ₀ is the scattering cross section, and Δτ is the time delay of the scattering point

The difference with τ, tri() represents the trigonometric function,

represents the antenna gain,

represents the distance from the transmitter to the scattering point,

is the distance from the receiver to the scattering point, Δf is the Doppler frequency of the scattering point

The difference from 0 Doppler, k represents the sampling times, B represents the equivalent noise bandwidth of the receiver, T _{rec_r} represents the equivalent input noise temperature of the down-view link, and T _{rec_d} represents the equivalent input noise of the up-view link temperature.

In the above method for improving the accuracy of sea surface measurement based on the coherent integration time optimization model, the following methods are used to obtain

和复功率波形的自相关性C_y(0,τ)；Obtain complex power waveform correlation

and the autocorrelation C _y (0,τ) of the complex power waveform;

From the incoherent mean N _I , the effective incoherent mean N _eff is obtained:

表示互相关波形之间的时间间隔；in,

represents the time interval between cross-correlated waveforms;

Expressing formula (7) with coherent integration time, we get:

表示上视天线热噪声的能量谱密度，

表示下视天线热噪声的能量谱密度；in,

represents the energy spectral density of the thermal noise of the upward looking antenna,

Represents the energy spectral density of the thermal noise of the downward looking antenna;

According to formula (8), we can get

包括：In the above method based on the coherent integration time optimization model to improve the accuracy of sea surface measurement, the correlation of the complex power waveform is obtained.

include:

次相干积分的复功率波形

determine the first

Subcoherently integrated complex power waveform

解算得到

According to y(nT _c ,τ) and

Solve to get

表示

的共轭轭。in,

express

the conjugate of .

In the above method for improving the accuracy of sea surface measurement based on the coherent integration time optimization model, σ _Z (τ) is determined as follows:

次相干积分的单次能量波形

determine the first

Single-shot energy waveform for subcoherent integration

解算得到单次能量波形之间的相关性

According to z(nT _c ,τ) and

Solve the correlation between single-shot energy waveforms

且

则有：Assuming that there is no coherence between the energy waveforms after incoherent averaging, that is:

and

Then there are:

表示时间间隔，Z(t,τ)表示能量波形在时间t的幅值；C_z(0,τ)表示当

时能量波形相关性的大小。in,

represents the time interval, Z(t, τ) represents the amplitude of the energy waveform at time t; C _z (0, τ) represents the time

The magnitude of the energy waveform correlation.

In the above-mentioned method for improving sea surface measurement accuracy based on the coherent integration time optimization model, the coherent integration time optimization model is constructed and obtained according to the estimated measurement accuracy model, including:

Substitute equations (5) and (9) into equation (1) to obtain the coherent integration time optimization model:

Correspondingly, the present invention also discloses a system for improving the high precision of sea surface measurement based on the coherent integration time optimization model, including:

The first model building module is used to convert the uncertainty of the energy waveform at the mirror point according to the transformation principle of the high-precision formula to obtain a model with high prediction and measurement accuracy;

The second model building module is used to construct and obtain the coherent integration time optimization model according to the pre-measured high-precision model;

The solving module is used to optimize the model according to the coherent integration time, respectively, to obtain the variation curve of the predicted measurement accuracy with the coherent integration time in different scenarios;

The result output module is used to filter out the minimum value points from the change curve, and use the estimated measurement accuracy corresponding to the minimum value point as the optimal measurement accuracy output.

The present invention has the following advantages:

The invention discloses a method for improving the high precision of sea surface measurement based on a coherent integration time optimization model. The coherent integration time optimization model is constructed based on the relationship between waveform correlation and coherent integration time. The airborne experimental data is used to verify the coherent integration time optimization model. The results show that the average deviation between the optimal measurement accuracy calculated based on the coherent integration time optimization model and the actual measurement accuracy is 0.16m; In the spaceborne simulation scenario, the coherent integration times corresponding to the optimal measurement accuracy are 7.5ms and 3.0ms, respectively. Compared with the empirical coherent integration processing, the measurement accuracy is improved by about 0.1m and 0.3m, respectively. The method for improving sea surface measurement accuracy based on the coherent integration time optimization model proposed by the present invention provides theoretical and method support for signal optimization processing and sea surface height inversion of GNSS-R altimetry verification satellites with high precision and high spatial resolution in the future .

Description of drawings

1 is a flowchart of a method for improving the high accuracy of sea surface measurement based on a coherent integration time optimization model in an embodiment of the present invention;

Fig. 2 is a kind of simplified intermediate frequency data processing flow chart in the embodiment of the present invention;

3 is an energy waveform diagram obtained by processing experimental data in an embodiment of the present invention;

4 is an energy waveform diagram obtained by a simulation in an embodiment of the present invention;

5 is a schematic diagram of a corresponding sea surface contribution area in an embodiment of the present invention;

Fig. 6 is a kind of comparison schematic diagram of the reciprocal value of the height measurement sensitivity of the measured data and the simulated data in the embodiment of the present invention;

7 is a schematic diagram of a comparison of an effective incoherent accumulation number and an incoherent accumulation number in an embodiment of the present invention;

8 is a schematic diagram of the number of effective incoherent accumulation times relative to mirror points in an embodiment of the present invention;

Fig. 9 is a schematic diagram of the actual measurement results when the coherent integration time is 10ms in the embodiment of the present invention; wherein, Fig. 9(a) is the height of the measured sea level relative to the WGS84 ellipsoid; The combined SSH residual;

10 is a schematic diagram of the variation curve of the measurement accuracy with the coherent integration time under a different situation in the embodiment of the present invention; wherein, the dotted line represents the actual measurement result, the square-dotted line represents the simulation accuracy considering the correlation between waveforms, and the triangular-dotted line Indicates the simulation accuracy without considering the correlation between waveforms;

FIG. 11 is a graph showing the variation of measurement accuracy with coherent integration time in an on-board scenario in an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the embodiments disclosed in the present invention will be described in further detail below with reference to the accompanying drawings.

It is of great significance for the application of GNSS-R satellite sea surface altimetry to improve the measurement accuracy under the requirement of ensuring the along-orbit resolution.

Improving the spatial resolution along the track will increase the correlation between the waveforms after signal processing, and a higher correlation will amplify the uncertainty of the waveform, thereby affecting the measurement accuracy. In order to obtain the optimal parameter combination of signal processing and then obtain the optimal solution of measurement accuracy, the invention discloses a method for improving sea surface measurement accuracy based on a coherent integration time optimization model, which is constructed based on the relationship between waveform correlation and coherent integration time. Coherent integration time optimization model. The airborne experimental data is used to verify the coherent integration time optimization model. The results show that the average deviation between the optimal measurement accuracy calculated based on the coherent integration time optimization model and the actual measurement accuracy is 0.16m; In the spaceborne simulation scenario, the coherent integration times corresponding to the optimal measurement accuracy are 7.5ms and 3.0ms, respectively. Compared with the empirical coherent integration processing, the measurement accuracy is improved by about 0.1m and 0.3m, respectively. The method for improving sea surface measurement accuracy based on the coherent integration time optimization model proposed by the present invention provides theoretical and method support for signal optimization processing and sea surface height inversion of GNSS-R altimetry verification satellites with high precision and high spatial resolution in the future .

As shown in Figure 1, in this embodiment, the method for improving the accuracy of sea surface measurement based on a coherent integration time optimization model includes:

Step 101: Convert the uncertainty of the energy waveform at the mirror point according to the conversion principle of the high-precision measurement formula, to obtain a model with high prediction and measurement accuracy.

In this embodiment, the expression of the obtained pre-measured high-precision model is as follows:

为Z(τ)的平均值，

为

的导数，

Among them, σ _h (τ) represents the pre-measurement accuracy, τ represents the time delay, c represents the speed of light in vacuum, i represents the incident angle, σ _Z (τ) represents the mean square error of the incoherent average energy waveform, _Sh (τ ) represents the altimetry sensitivity, Z(τ) represents the incoherent average energy waveform,

is the mean value of Z(τ),

for

the derivative of ,

和

的求解过程如下：preferably,

and

The solution process is as follows:

The direct signal and the reflected signal are represented by the signal component and the noise component, and the coherent integration process is performed to obtain the complex power waveform y(nT _c ,τ) of the nth coherent integration:

y(nT _c ,τ)=y _s (nT _c ,τ)+y _nd (nT _c ,τ)+y _nr (nT _c ,τ)+y _ndr (nT _c ,τ)…(2)

According to the obtained y(nT _c ,τ), determine the single energy waveform z(nT _c ,τ) of the nth coherent integration:

z(nT _c ,τ)=y(nT _c ,τ)y ^* (nT _c ,τ)…(3)

According to z(nT _c ,τ), the incoherent average energy waveform Z(τ) is obtained by solving:

Simultaneous equations (2), (3) and (4) yield Z(τ) in terms of coherent integration time:

Z(τ)=<|y _s (nT _c ,τ)| ² >+<|y _nr (nT _c ,τ)| ² >+<|y _nd (nT _c ,τ)| ² >+<|y _ndr (nT _c ,τ)| ² >…(5)

对

求导，得到

According to formula (5), average Z(τ) to get

right

seek guidance, get

Among them, n represents the coherent integration times, T _c represents the coherent integration time, nT _c represents the process of the nth coherent integration, y _s (nT _c , τ) represents the cross-correlation energy value of the useful signal term, y _nd (nT _c , τ) represents the cross-correlation energy value of the noise term of the direct signal and the reflected signal, y _nr (nT _c ,τ) represents the cross-correlation energy value of the noise term of the direct signal and the reflected signal, and y _ndr (nT _c ,τ) represents the noise of the direct signal The cross-correlation energy value of the term and the reflected signal noise term, y ^* (nT _c , τ) represents the conjugate of y(nT _c , τ), _NI represents the incoherent average, and <> represents the average value.

Further, the coherent integration time expressions of y _s (nT _c ,τ), y _nd (nT _c ,τ), y _nr (nT _c ,τ) and y _ndr (nT _c ,τ) are as follows:

表示散射点向量，p表示散射点面积，σ₀表示散射截面，Δτ表示散射点时延

与τ的差值，tri()表示三角函数，

表示天线增益，

表示发射机到散射点的距离，

表示接收机到散射点的距离，Δf表示散射点的多普勒频率

与0多普勒的差值，k表示采样次数，B表示接收机的等效噪声带宽，T_{rec_r}表示下视链路的等效输入噪声温度，T_{rec_d}表示上视链路的等效输入噪声温度。Among them, P _d is the total energy of the direct signal at the input of the correlator, P _t is the energy of the transmitted signal, P _r is the total energy of the reflected signal at the input of the correlator,

is the scattering point vector, p is the area of the scattering point, σ ₀ is the scattering cross section, and Δτ is the time delay of the scattering point

The difference with τ, tri() represents the trigonometric function,

represents the antenna gain,

represents the distance from the transmitter to the scattering point,

is the distance from the receiver to the scattering point, Δf is the Doppler frequency of the scattering point

The difference from 0 Doppler, k represents the sampling times, B represents the equivalent noise bandwidth of the receiver, T _{rec_r} represents the equivalent input noise temperature of the down-view link, and T _{rec_d} represents the equivalent input noise of the up-view link temperature.

的求解过程如下：preferably,

The solution process is as follows:

和复功率波形的自相关性C_y(0,τ)；Obtain complex power waveform correlation

and the autocorrelation C _y (0,τ) of the complex power waveform;

From the incoherent mean N _I , the effective incoherent mean N _eff is obtained:

Expressing formula (7) with coherent integration time, we get:

According to formula (8), we can get

表示互相关波形之间的时间间隔，

表示上视天线热噪声的能量谱密度，

表示下视天线热噪声的能量谱密度。in,

represents the time interval between the cross-correlated waveforms,

represents the energy spectral density of the thermal noise of the upward looking antenna,

Represents the energy spectral density of the downward looking antenna thermal noise.

的求解过程如下：further,

The solution process is as follows:

次相干积分的复功率波形

determine the first

Subcoherently integrated complex power waveform

解算得到

According to y(nT _c ,τ) and

Solve to get

表示

的共轭轭。in,

express

the conjugate of .

Preferably, the solution process of σ _Z (τ) is as follows:

次相干积分的单次能量波形

determine the first

Single-shot energy waveform for subcoherent integration

解算得到单次能量波形之间的相关性

According to z(nT _c ,τ) and

Solve the correlation between single-shot energy waveforms

且

则有：Assuming that there is no coherence between the energy waveforms after incoherent averaging, that is:

and

Then there are:

表示时间间隔，Z(t,τ)表示能量波形在时间t的幅值；C_z(0,τ)表示当

时能量波形相关性的大小。in,

represents the time interval, Z(t, τ) represents the amplitude of the energy waveform at time t; C _z (0, τ) represents the time

The magnitude of the energy waveform correlation.

Step 102 , construct and obtain the coherent integration time optimization model according to the pre-measured high-precision model.

In this embodiment, the above equations (5) and (9) are substituted into equation (1) to obtain the coherent integration time optimization model:

Step 103 , according to the coherent integration time optimization model, respectively calculate and obtain the variation curve of the pre-measured high precision with the coherent integration time in different scenarios.

In step 104, a minimum value point is obtained by screening from the change curve, and the estimated measurement precision corresponding to the minimum value point is used as the optimal measurement precision output.

On the basis of the above-mentioned embodiment, the following description will be given with reference to a specific example.

1. Data processing and altimetry inversion

In order to verify the accuracy of the coherent integration time optimization model, the embodiments of the present invention are verified by using the data obtained by the Spanish IEEC over the Baltic Sea through airborne experiments on December 3, 2015.

During the experiment, the height of the receiver is about 3km and the speed is about 50m/s. The top-view antenna and the bottom-view antenna mounted on the aircraft each have 8 antenna units, which can receive direct and reflected signals at the same time. The radio frequency signal received by the antenna unit is filtered, amplified and down-converted into an intermediate frequency signal. The intermediate frequency signal is quantized by a comparator, and the quantized signal is connected to a Field Programmable Gate Array (FPGA) through a D flip-flop, and its parallel processing capability is used to sample the signal. Finally, the sampled digital signal is transmitted to the ground receiving station in real time through the PCIe bus. At each sampling, the analog signal of each antenna unit is quantized into 1-bit in-phase component and 1-bit quadrature component, so each sampling of 16 antenna units will output a 32-bit data stream. The FPGA sampling frequency is 80MHz, so the size of the original data set sampled per second is 320MB. The raw data collected by the FPGA every second will be divided into 4882×64KB data blocks and transmitted to the ground, each data block occupies 4KB space to store auxiliary information (such as time stamp), so the actual size of the raw data available per second is 319,927,224KB .

The acquired data is processed by a software-defined receiver. Figure 2 shows a simplified flow chart of intermediate frequency data processing and altimetry inversion. The specific processing steps are as follows:

(1) Data reading: read 4 bytes of binary data each time, corresponding to the in-phase component and quadrature component of 16 antenna units, and convert the binary data from 0, 1 encoding to 1, -1 encoding (that is, logic electrical The level is converted to the signal level of non-return-to-zero polarity), and the direct signal and reflected signal of each sample can be expressed as:

分别对应直射信号与反射信号在第n次相干积分时的第k个采样值的大小，s表示每个天线单元的信号分量，i对应天线单元号，up_I、up_Q分别表示上视天线单元的同相分量与正交分量，down_I、down_Q分别表示下视天线单元的同相分量与正交分量。in,

Corresponding to the size of the k-th sampling value of the direct signal and the reflected signal in the n-th coherent integration, s represents the signal component of each antenna element, i corresponds to the antenna element number, up_I, up_Q respectively represent the in-phase of the up-view antenna element The component and the quadrature component, down_I and down_Q respectively represent the in-phase component and the quadrature component of the down-view antenna unit.

(2) Coherent integration: Set the coherent integration time as T _c and the sampling frequency of the signal as f _s =80MHz, then the sequence of each coherent integration is composed of f _s ×T _c sampling points, namely:

转换到频域

在频谱中找到幅值异常的单频干扰并将其滤掉。滤波后的直射信号与反射信号在频域进行互相关，得到复功率波形，即：In order to prevent the single-frequency noise interference from affecting the final data processing result, the single-frequency noise interference needs to be removed before coherent integration. The method is to convert the time domain

Convert to frequency domain

Find and filter out single-frequency interference with abnormal amplitude in the spectrum. The filtered direct signal and the reflected signal are cross-correlated in the frequency domain to obtain a complex power waveform, namely:

It should be stated that the phased array antenna is used in the airborne experiment, and beamforming is usually used to improve the signal-to-noise ratio in the data post-processing process. However, the effect of beamforming can drown out the effect of coherent integration time on the waveform signal-to-noise ratio. In order to amplify the influence of the coherent integration time on the measurement accuracy, the present invention does not use beamforming processing.

(3) Re-tracking and incoherent accumulation: The power waveform after coherent integration needs to be incoherently accumulated to reduce the influence of speckle noise. Considering the drift motion of the aircraft in the vertical direction during the accumulation process, the waveform delay needs to be compensated (re-tracking) in each incoherent accumulation process. The waveform after incoherent accumulation can be represented by formula (4), and Fig. 3 shows the energy waveform obtained after incoherent accumulation.

(4) Delay estimation and error correction: The mirror point delay of the reflected signal can be estimated from the processed waveform. The mirror point tracking method is divided into two methods: fixed point tracking and waveform fitting. Fixed-point tracking is divided into three methods: DER, MAX, and HALF. For the sake of simplicity, this time, the mirror point delay is estimated by DER, which can be expressed as

表示从实测波形(图3)中计算的导数最大值点对应的时延，

对应仿真波形的镜面点时延与导数最大值点时延。仿真波形是基于Z-V模型生成的，其考虑了几何、仪器配置和海洋状态的影响，已被广泛应用于GNSS-R中，图4表示对应仿真得到的时延能量波形图。in,

represents the time delay corresponding to the derivative maximum point calculated from the measured waveform (Fig. 3),

Corresponding to the mirror point delay and the derivative maximum point delay of the simulated waveform. The simulation waveform is generated based on the ZV model, which takes into account the influence of geometry, instrument configuration and ocean state, and has been widely used in GNSS-R. Figure 4 shows the corresponding simulated time-delay energy waveform.

The estimated mirror point delay information also needs to be corrected for errors, including the tropospheric delay distance error ρ _trop , the ionospheric delay distance error p _iono , and the sky baseline attitude error p _bl . Since the ionosphere is located in the space above 60km, the direct and reflected signals received by the airborne platform have experienced the same downlink transmission path, and the ionospheric errors of the direct and reflected signals are almost the same, so the delay deviation caused by the ionosphere can be ignored. The tropospheric delay can be estimated by a simple mathematical model:

Among them, e is the satellite elevation angle, HR is the receiver height, and _{H trop} is the average height of the troposphere. The sky baseline error is obtained by the path difference between the up-looking antenna and the down-looking antenna relative to the mirror point under the known airborne position and attitude information. The corrected specular point delay can be converted to the true height value from the receiver to the sea surface:

(5) Height inversion and accuracy calculation: After time delay estimation and error correction, the vertical distance from the sea level of the measured area to the reference ellipsoid can be calculated:

表示利用双基几何求得的接收机在WGS84参考椭球面的高度值。为了评估热噪声与散斑噪声引起的信号随机性对高度随机误差的影响，用拟合的分段线性函数减去测量到的SSH序列，得到零均值、接近白噪声的SSH残差序列。计算SSH残差序列的标准差作为测高精度

采用线性拟合而不是大地水准面模型是为了避免模型误差的影响。in,

Indicates the height value of the receiver on the WGS84 reference ellipsoid obtained by using the bibasic geometry. To evaluate the effect of signal randomness caused by thermal and speckle noise on highly random errors, the measured SSH sequence was subtracted from the fitted piecewise linear function to obtain a zero-mean, near-white noise SSH residual sequence. Calculate the standard deviation of the SSH residual series as the measurement accuracy

A linear fit rather than a geoid model is used to avoid the effects of model errors.

2. Construction of the coherent integration time optimization model

The relative delay between the direct signal and the reflected signal needs to be extracted from the energy waveform. In the process of signal processing, the reflected signal and the local C/A code replica (cGNSS-R) or the direct signal (iGNSS-R) are cross-correlated in the frequency domain to obtain a set of complex powers containing in-phase and quadrature components. waveform. In this embodiment of the present invention, iGNSS-R is taken as an example, and it is assumed that the coherent integration time is T _c , then the complex power waveform after coherent integration can be represented by a discrete array y(nT _c ,τ):

y(nT _c ,τ)=[y(nT _c ,τ ₁ ),y(nT _c ,τ ₂ )...y(nT _c ,τ _n )]...(21)

Among them, τ _n represents the time delay corresponding to the nth coherent integration, and the time delay resolution is inversely proportional to the signal sampling rate f _s . The square of the complex power waveform yields the single-shot energy waveform expressed as:

z(nT _c ,τ)=y(nT _c ,τ)y ^* (nT _c ,τ)…(3)

Assuming that the incoherent average is N _I , the incoherent average energy waveform Z(τ) can be expressed as:

Wherein, T _I =T _c ×N _I .

的复功率波形的相关性是由来自两个不同海表面信号的相干性决定。这些表面区域呈椭圆状，椭圆半轴的长度与τ_n有关。复功率波形相关性

的大小定义为：Due to the relative motion of the transmitter and receiver, the two times are separated

The correlation of the complex power waveform is determined by the coherence of the signals from two different sea surfaces. These surface areas are elliptical, and the length of the semi-axis of the ellipse is related to τ _n . Complex Power Waveform Correlation

The size of is defined as:

变化时由相关函数定义的几何图形，随着

的增大，两个海表面散射信号之间的相关性逐渐变小。Figure 5, illustrates when the parameters

The geometry defined by the correlation function when changing, with

increases, the correlation between the two sea surface scattering signals gradually becomes smaller.

定义为：Correlation between single-shot energy waveforms

defined as:

且

则非相干平均后能量波形的均方差σ_Z(τ)可以表示为：Because the time of incoherent averaging is long, it can be assumed that there is no coherence between the energy waveforms after incoherent averaging, that is,

and

Then the mean square error σ _Z (τ) of the energy waveform after incoherent averaging can be expressed as:

According to the conversion principle of the measurement precision formula, the estimated measurement precision can be converted by the uncertainty of the energy waveform at the mirror point, namely:

表示标准差与能量的比值，由于考虑了波形相关性的影响，该比值

表示为

而非

Represents the ratio of standard deviation to energy, which takes into account the influence of waveform correlation

Expressed as

instead of

The first item in the above formula (1) is only determined by the observation geometry, and the altimetry sensitivity and the number of effective waveforms are complex functions related to the coherent integration time, which will be analyzed in detail next.

2.1) Height measurement sensitivity

The received direct signal and reflected signal can be represented by signal components and noise components, and the complex power waveform y(nT _c ,τ) after coherent integration of the two can be represented as:

y(nT _c ,τ)=y _s (nT _c ,τ)+y _nd (nT _c ,τ)+y _nr (nT _c ,τ)+y _ndr (nT _c ,τ)…(2)

Assuming that there is no coherence between the signal component and the noise component of the waveform, the energy waveform Z(τ) after incoherent averaging can be expressed as:

Z(τ)=<|y _s (nT _c ,τ)| ² >+<|y _nr (nT _c ,τ)| ² >+<|y _nd (nT _c ,τ)| ² >+<|y _ndr (nT _c ,τ)| ² >…(5)

y _s (nT _c ,τ), y _nd (nT _c ,τ), y _nr (nT _c ,τ) and y _ndr (nT _c ,τ) can be replaced by coherent integration times:

Based on Equation (5) and Equation (6), the magnitude of the altimetry sensitivity can be represented by the coherent integration time. Figure 6 compares the variation curve of the reciprocal of the measured sensitivity and the simulated altimetry sensitivity with the coherent integration time, which is obtained according to the second section. Considering the requirement of high along-track spatial resolution, the total processing time of one energy waveform is set to T _I =1s. It can be seen that the curve of the measured data and the simulated data are in good agreement, and the inverse of the sensitivity will decrease with the increase of the coherent integration time and approach a limit value. Differences between the curves may be caused by noise terms that are not considered.

2.2) Number of valid waveforms

波形之间相关性的大小可以用公式(10)来表示，那么有效波形数的倒数可以进一步推导为：Since the influence of correlation between waveforms is considered, N _eff is used to replace _NI to represent the ratio of energy to standard deviation, namely:

The magnitude of the correlation between waveforms can be expressed by formula (10), then the reciprocal of the number of effective waveforms can be further deduced as:

Among them, the specific derivation process of formula (7) is as follows:

可以推导为：Simultaneously formula (3) and formula (11),

It can be deduced as:

Usually, the complex waveform strictly follows the circular complex Gaussian statistics, and the correlation function of the primary energy waveform can be simplified using the complex Gaussian moment theorem:

Substitute equation (23) into equation (22):

According to Equation (4) and Equation (12), the variance of the waveform energy after incoherent averaging can be expressed as:

Substituting equation (4) into the first term of equation (26) yields:

According to formula (22), <z(iT _c ,τ)z(jT _c ,τ)> can be replaced by:

Substituting Equation (27) and Equation (28) into Equation (26), we get:

Simultaneously formula (24), formula (25) and formula (29) can be derived:

The formula (30) is the formula (7).

Assuming that the signal is uncorrelated with noise, N _eff can be expressed in terms of coherent integration time as:

Among them, the specific derivation process of formula (8) is as follows:

Assuming that there is no correlation between the complex waveform signal and noise, the covariance function for the iGNSS-R waveform can contain the following four terms:

Since the reflected signal noise is much larger than the direct signal, it can be assumed that _Cy,nu (nT _c ,τ)≈0. The covariance of the signal components can be simplified to:

The covariance of the noise component can be simplified to:

Simultaneously formula (31), formula (32) and formula (33) can be obtained:

By combining formula (30) and formula (34), the relationship between the inverse of the effective incoherent mean and the coherent integration time can be derived to obtain formula (8).

那么N_eff＝N_I。如图8所示，N_eff也会随着不同的时延cτ_n而变化，这是信号分量和多普勒带宽的混合效应。It can be seen that _Neff is a complex function of coherent integration time. Compared with the traditional way of directly using N _I to solve the variance of the average waveform, N _eff takes into account the correlation between waveforms, so that the variance of the incoherently averaged waveform calculated within the same integration time is higher than the value of the traditional method. is larger, and the predicted measurement accuracy value is larger. Based on formula (8), the relationship between N _eff , _NI and coherent integration time simulated in the airborne scenario is shown in Fig. 7 . As can be seen from Figure 7, when the integration time T _I is a constant, N _I and T _c have an inversely proportional relationship, and N _eff and T _c have a nonlinear relationship. The difference between N _eff and N _I will gradually decrease as the coherent integration time increases. This change is reasonable. According to formula (7), as the coherent integration time increases, the correlation between the waveforms becomes smaller, and N _eff gradually approaches _NI . If there is no correlation between waveforms,

Then N _eff =N _I . As shown in Fig. 8, N _eff also varies with different time delays cτ _n , which is a mixed effect of signal components and Doppler bandwidth.

It can be seen from Fig. 6 that a larger coherent integration time is desired in order to improve the height measurement sensitivity and thus the measurement accuracy. According to Fig. 7, it is desirable that the coherent integration time be a small value as the number of effective waveforms restricts the measurement accuracy. Therefore, an optimal measurement accuracy result requires a compromise between the coherent integration time.

3. Results and Discussion

3.1) Validation of coherent integration time optimization model

In order to verify the validity of the coherent integration time optimization model, the embodiment of the present invention compares the measurement accuracy results obtained according to the coherent integration time optimization model formula (1) with the airborne experimental data measurement accuracy results.

The calculation method of the measured accuracy is: keep the waveform processing time T _I unchanged, adjust the coherent integration time T _c , repeat the process of "data processing and altimetry inversion", and obtain the SSH residual with the coherent integration time T _c as the independent variable. difference series and calculate the size of the measured high precision value. In order to avoid the error interference caused by the aircraft during the turning process, the present invention selects the straight flight trajectory to calculate and measure the high precision. Figure 9(a) corresponds to the calculated SSH sequences during two straight flight trajectories (intraday seconds correspond to 39102-39521 and 39942-40721. The total processing time of one waveform is 1s, and the coherent integration time corresponds to 10ms), SSH residuals The sequence is shown in Figure 9(b).

The calculation method of the accuracy of the coherent integration time optimization model is as follows: (1) Obtain auxiliary information data during the experiment, including: receiver position and velocity information after dual-frequency orbit determination; Position speed information; wind speed information during the experiment; load parameter information. (2) Calculate the incident angle based on the Z-V model, and calculate the change sequence of the height measurement sensitivity and the number of effective waveforms with the coherent integration time according to formula (5), formula (6) and formula (8) respectively, and finally substitute the results into the formula ( 1) The variation curve of model accuracy with coherent integration time is obtained.

Figure 10 shows the variation of accuracy with coherent integration time under different conditions. In order to evaluate the consistency between the curves, the average deviation of the accuracy <|p ^m -p ^obs |> is taken as the evaluation standard, where p ^m is the simulation accuracy and p ^obs is the measured accuracy. The results show that the average deviation between the measured accuracy and the model accuracy without waveform correlation is 0.85m, and the average deviation between the measured accuracy and the model accuracy without waveform correlation is 0.16m. It can be seen that the simulation accuracy considering the waveform correlation is in good agreement with the measured results.

There is still a certain deviation between the coherent integration time optimization model curve and the measured curve. On the one hand, the measured accuracy results are generally worse than the simulation results, with an average deviation of 0.16m. This may be due to the reduction of the signal-to-noise ratio due to the lack of beamforming processing during the waveform processing. On the other hand, the variation trend of the measured accuracy curve shows a certain volatility. This is because, different from the simulation results, the actual measurement may contain other random deviations that are not considered, such as the mechanical vibration of the aircraft, attitude changes, and the orientation of the antenna. These random factors can be used to optimize the model in the future by obtaining the recorded data of aircraft attitude and vibration.

3.2) Application of coherent integration time optimization model

Airborne scene

The consistency between the model and the measured results indicates that the proposed model can better reflect the influence of different parameter processing on the measurement accuracy. In this regard, the model can be applied to different altimetry scenarios to provide a reference for the optimization of intermediate frequency data.

In the airborne experimental scenario, the coherent integration time corresponding to the optimal accuracy is estimated according to the simulation model curve. The corresponding measured data is processed based on the coherent integration time solved by the model. It can be seen from Figure 10 that the optimal coherent integration time corresponding to the model curve is 7.5ms. The optimal coherent integration time obtained by the model is used to process the measured data, and the accuracy value is 0.81m. Compared with the empirical coherent integration processing (10ms), the measurement accuracy value is improved by 0.09m.

Onboard scene

Considering the needs of future spaceborne altimetry tasks, the proposed model was applied to the spaceborne altimetry scene to predict the optimal data processing parameters. The processing time of the energy waveform was set to 1 s.

Figure 11 shows the variation curve of the measured high precision with the coherent integration time in the on-board scenario obtained from the simulation. It can be seen from Figure 11 that the optimal coherent integration time in this scenario is 3ms, and the corresponding measurement accuracy is 0.35m. Compared with the traditional on-board coherent integration time (1ms), the measurement accuracy is increased by 0.31m. The spatial region corresponding to each signal processing is the result of the joint action of the equal delay circle and the equal Doppler curve. Compared with the airborne receiver, the spaceborne receiver is faster, the coincidence rate of the equal delay rings corresponding to adjacent waveforms is small, and the Doppler bandwidth is wider, which makes the coherence time of the reflected signal shorter, and the effective incoherent accumulation number Become more. As a result, the optimal coherent integration time of the spaceborne is smaller than that of the airborne, and the measurement accuracy value is higher.

On the basis of the above-mentioned embodiment, the present invention also discloses a system for improving the high precision of sea surface measurement based on the coherent integration time optimization model, including: a first model building module, which is used for transforming the mirror surface points according to the transformation principle of the measurement precision formula. The uncertainty of the energy waveform at the location is converted to obtain a high-precision prediction model; the second model building module is used to construct a coherent integration time optimization model according to the high-precision prediction model; The integration time optimization model is used to calculate the variation curve of the estimated high precision with the coherent integration time in different scenarios; the result output module is used to select the minimum value points from the change curve, and the corresponding prediction value of the minimum value points is obtained. The measurement accuracy is used as the output of the optimal measurement accuracy.

As for the system embodiment, since it corresponds to the method embodiment, the description is relatively simple, and for related parts, please refer to the description of the method embodiment part.

Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can use the methods and technical contents disclosed above to improve the present invention without departing from the spirit and scope of the present invention. The technical solutions are subject to possible changes and modifications. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solutions of the present invention belong to the technical solutions of the present invention. protected range.

Contents that are not described in detail in the specification of the present invention belong to the well-known technology of those skilled in the art.

Claims (10)

1. a method for improving sea surface measurement accuracy based on coherent integration time optimization model, is characterized in that, comprises: According to the transformation principle of the measurement high-precision formula, the uncertainty of the energy waveform at the mirror point is converted to obtain the pre-measured high-precision model; According to the pre-measured high-precision model, the coherent integration time optimization model is constructed and obtained; According to the coherent integration time optimization model, the variation curves of the pre-measured high precision with the coherent integration time under different scenarios are calculated respectively; The minimum value points are obtained from the change curve, and the estimated measurement accuracy corresponding to the minimum value point is used as the optimal measurement accuracy output. 2. the method for improving sea surface measurement high precision based on coherent integration time optimization model according to claim 1, is characterized in that, the expression of the obtained pre-estimation measurement high precision model is as follows:

Among them, σ _h (τ) represents the pre-measurement accuracy, τ represents the time delay, c represents the speed of light in vacuum, i represents the incident angle, σ _Z (τ) represents the mean square error of the incoherent average energy waveform, _Sh (τ ) represents the altimetry sensitivity, Z(τ) represents the incoherent average energy waveform,

is the mean value of Z(τ),

for

derivative of . 3. the method for improving sea surface measurement accuracy based on coherent integration time optimization model according to claim 2, is characterized in that, the solution formula of height measurement sensitivity S _h (τ) is as follows:

4. the method for improving sea surface measurement accuracy based on coherent integration time optimization model according to claim 3, is characterized in that, is determined by the following way

and

The direct signal and the reflected signal are represented by the signal component and the noise component, and the coherent integration process is performed to obtain the complex power waveform y(nT _c ,τ) of the nth coherent integration: y(nT _c ,τ)=y _s (nT _c ,τ)+y _nd (nT _c ,τ)+y _nr (nT _c ,τ)+y _ndr (nT _c ,τ)…(2) Among them, n represents the coherent integration times, T _c represents the coherent integration time, nT _c represents the process of the nth coherent integration, y _s (nT _c , τ) represents the cross-correlation energy value of the useful signal term, y _nd (nT _c , τ) represents the cross-correlation energy value of the noise term of the direct signal and the reflected signal, y _nr (nT _c ,τ) represents the cross-correlation energy value of the noise term of the direct signal and the reflected signal, and y _ndr (nT _c ,τ) represents the noise of the direct signal The cross-correlation energy value of the term and the reflected signal noise term; According to the obtained y(nT _c ,τ), determine the single energy waveform z(nT _c ,τ) of the nth coherent integration: z(nT _c ,τ)=y(nT _c ,τ)y ^* (nT _c ,τ)…(3) Among them, y ^* (nT _c ,τ) represents the conjugate of y(nT _c ,τ); According to z(nT _c ,τ), the incoherent average energy waveform Z(τ) is obtained by solving:

Among them, _NI represents the incoherent average, and <> represents the average value; Simultaneous equations (2), (3) and (4) yield Z(τ) in terms of coherent integration time: Z(τ)=<|y _s (nT _c ,τ)| ² >+<|y _nr (nT _c ,τ)| ² >+<|y _nd (nT _c ,τ)| ² >+<|y _ndr (nT _c ,τ)| ² >…(5) According to formula (5), average Z(τ) to get

right

seek guidance, get

5. the method for improving sea surface measurement accuracy based on coherent integration time optimization model according to claim 4, is characterized in that, y _s (nT _c , τ), y _nd (nT _c , τ), y _nr (nT _c , τ) ,τ) and y _ndr (nT _c ,τ) are expressed as coherent integration times as follows:

Among them, P _d is the total energy of the direct signal at the input of the correlator, P _t is the energy of the transmitted signal, P _r is the total energy of the reflected signal at the input of the correlator,

is the scattering point vector, p is the area of the scattering point, σ ₀ is the scattering cross section, and Δτ is the time delay of the scattering point

The difference with τ, tri() represents the trigonometric function,

represents the antenna gain,

represents the distance from the transmitter to the scattering point,

is the distance from the receiver to the scattering point, Δf is the Doppler frequency of the scattering point

The difference from 0 Doppler, k represents the sampling times, B represents the equivalent noise bandwidth of the receiver, T _{rec_r} represents the equivalent input noise temperature of the down-view link, and T _{rec_d} represents the equivalent input noise of the up-view link temperature. 6. the method for improving the high precision of sea surface measurement based on the coherent integration time optimization model according to claim 5, is characterized in that, obtains by the following method

Obtain complex power waveform correlation

and the autocorrelation C _y (0,τ) of the complex power waveform; From the incoherent mean N _I , the effective incoherent mean N _eff is obtained:

in,

represents the time interval between cross-correlated waveforms; Expressing formula (7) with coherent integration time, we get:

in,

represents the energy spectral density of the thermal noise of the upward looking antenna,

Represents the energy spectral density of the thermal noise of the downward looking antenna; According to formula (8), we can get

7. the method for improving sea surface measurement accuracy based on coherent integration time optimization model according to claim 6, is characterized in that, obtains complex power waveform correlation

include: determine the first

Subcoherently integrated complex power waveform

According to y(nT _c ,τ) and

Solve to get

in,

express

the conjugate of . 8. the method for improving sea surface measurement accuracy based on coherent integration time optimization model according to claim 7, is characterized in that, by the following manner, determine σ _Z (τ): determine the first

Single-shot energy waveform for subcoherent integration

According to z(nT _c ,τ) and

Solve the correlation between single-shot energy waveforms

Assuming that there is no coherence between the energy waveforms after incoherent averaging, that is:

and

Then there are:

in,

represents the time interval, Z(t, τ) represents the amplitude of the energy waveform at time t; C _z (0, τ) represents the time

The magnitude of the energy waveform correlation. 9. the method for improving the high accuracy of sea surface measurement based on the coherent integration time optimization model according to claim 8, it is characterized in that, according to the pre-estimation measurement high precision model, the construction obtains the coherent integration time optimization model, comprising: Substitute equations (5) and (9) into equation (1) to obtain the coherent integration time optimization model:

10. A system for improving sea surface measurement accuracy based on a coherent integration time optimization model, characterized in that it comprises: The first model building module is used to convert the uncertainty of the energy waveform at the mirror point according to the transformation principle of the high-precision formula to obtain a model with high prediction and measurement accuracy; The second model building module is used to construct and obtain the coherent integration time optimization model according to the pre-measured high-precision model; The solving module is used to optimize the model according to the coherent integration time, respectively, to obtain the variation curve of the predicted measurement accuracy with the coherent integration time in different scenarios; The result output module is used to filter out the minimum value points from the change curve, and use the estimated measurement accuracy corresponding to the minimum value point as the optimal measurement accuracy output.

Images