Sea ice detection method based on GNSS-IR power ratio data

Abstract

The invention discloses a sea ice detection method based on GNSS-IR power ratio data, and relates to the technical field of sea ice detection. The method specifically comprises the following steps: extracting a direct power sequence and an interference signal sequence from an SNR (Signal to Noise Ratio) through a vector relationship among a direct signal, a reflected signal and an interference signal, and carrying out Hilbert transform on the interference signal to obtain a reflected signal power sequence; and an average is taken according to different frequency points of different satellites and is used as a daily average power ratio factor, so that occurrence and ablation of sea ice are detected by using a time change sequence of a power ratio. The method is different from a traditional ice period detection method, sea ice can be more independently and directly detected, the situation that a GNSS-IR traditional model is poor in independence is improved, and sea ice changes can be more obviously detected.

Description

Technical Field

[0001] The present invention relates to the technical field of sea ice detection, and in particular to a method for realizing sea ice detection based on GNSS-IR power ratio data. Background Art

[0002] Sea ice is an important part of the cryosphere. The monitoring of sea ice has a significant impact on global climate change. In addition, the presence or absence of sea ice is crucial to maritime transportation, marine fisheries, coastal port operations, and the development of marine oil and gas resources. Traditional sea ice detection methods mainly use visible light remote sensing and microwave remote sensing. Since the L-band microwave signal used by GNSS-R has different reflection signal intensities for sea ice and seawater, it can also be used for sea ice detection. At present, there are three main methods of sea ice detection that are widely used in practice: the first method is to detect sea ice based on medium and low resolution optical sensors, which can obtain more intuitive images, but is often limited by natural weather such as clouds and fog, and cannot guarantee the real-time and continuity of sea ice monitoring. The second method is mainly to detect sea ice through SAR (synthetic aperture radar). Although it can ensure real-time imaging, it requires high costs when conducting large-scale sea ice monitoring. The third method is to use dual-antenna GNSS-R technology, which uses a special receiver and receives direct signals and reflected signals through two antennas respectively. Compared with the first two methods, the third method is low-cost, but it requires a special receiver to process the data. The GNSS-IR technology used in the present invention was proposed by Larson (Larson, 2008), which uses traditional geodetic receivers and antennas to process the multipath signals in the observation files to obtain the inversion results. Compared with other methods, it is easier to implement and has lower costs. Summary of the invention

[0003] The purpose of the present invention is to propose a sea ice detection method based on GNSS-IR power ratio observation, which is simple, easy to operate and low-cost, and provides an effective method for monitoring sea ice changes using GNSS-IR technology.

[0004] In order to achieve the above object, the present invention adopts the following technical scheme: A sea ice detection method based on GNSS-IR power ratio data, comprising: Through the vector relationship between the direct signal, reflected signal and interference signal, the direct power sequence and the interference signal sequence are extracted from the SNR, and the interference signal is Hilbert transformed to obtain the reflected signal power sequence. Then, the reflected / direct signal power ratio sequence is obtained and averaged according to the different frequencies of different satellites as the daily average power ratio factor, so as to use the time variation sequence of the power ratio to detect the appearance and melting of sea ice.

[0005] Furthermore, the direct power sequence and the interference signal sequence are extracted from the SNR through the vector relationship between the direct signal, the reflected signal and the interference signal, specifically: When the reflected signal and the direct signal are received by the same antenna at the same time, an interference effect will occur. The power of this interference effect is measured by the signal-to-noise ratio recorded by the receiver. The mathematical expression of the signal-to-noise ratio is: ; in, , represent the direct signal amplitude and the reflected amplitude respectively; It represents the phase difference of the reflected signal relative to the direct signal. is the noise power; the vertical distance from the antenna phase center to the reflecting surface is h, and the phase delay of the reflected signal compared to the direct signal is It is expressed as: ; in: is the satellite altitude angle, is the signal wavelength; The trend term in SNR is obtained by using the third-order polynomial fitting method , which is the direct signal power; use SNR to subtract the fitted direct signal component , thereby separating the interference signal component, that is, the residual signal-to-noise ratio dSNR; and because the geometric characteristics of the reflecting surface affect the roughness of the reflecting surface ,in is the root mean square of the height of the reflecting surface, which in turn causes the attenuation of the reflected signal as the altitude angle changes , which can be expressed as: ; So the residual signal-to-noise ratio dSNR can finally be expressed as: ; in , the residual signal-to-noise ratio dSNR obtained is the interference signal sequence.

[0006] Furthermore, the interference signal is subjected to Hilbert transform to obtain the reflected signal power sequence, specifically: The residual signal-to-noise ratio sequence dSNR is processed using the Hilbert transform, and the transformed signal is expressed as: ; in represents the residual signal-to-noise ratio sequence dSNR, is the Hilbert transform symbol, is the integration variable.

[0007] Construct an analytic signal related to the real signal in the time domain : ; The real part is the original reflected signal, and the imaginary part is the signal after Hilbert transformation; combined with Euler's formula, we finally get: ; in , , is the load wave signal, is the complex envelope; for the analytical signal The absolute value can be obtained to obtain the envelope of the analytical signal, that is, the amplitude: ; The amplitude sequence thus obtained is taken as the reflected signal power .

[0008] Furthermore, the reflected / direct signal power ratio sequence is obtained and averaged according to different frequency points of different satellites as the daily average power ratio factor, so as to detect the appearance and melting of sea ice using the time variation sequence of the power ratio, specifically: Power ratio sequence of reflected signal to direct signal It is expressed as: ; According to the frequency of different frequency points, the reflected / direct power ratio of satellite signals of the same frequency and different systems is averaged every day as the daily average power ratio factor. Long-term observation forms a time series, so as to detect the appearance and melting of sea ice according to the changes in the value of the time series factor.

[0009] The beneficial effects of the present invention are embodied in: This method is not affected by natural weather such as clouds and fog, ensuring the real-time and continuity of sea ice detection. It only requires one antenna and does not require a special receiver, so it is low-cost and convenient for data collection. This method uses the interference signal and its vector relationship with the direct and reflected components to establish a geometric model. The direct signal power is obtained by fitting the SNR sequence with a third-order polynomial, and the reflected signal power is obtained by performing a Hilbert transform on the interference signal. The reflected / direct power ratio sequence is solved and the average of different frequency points of different satellite systems is taken as the daily average power ratio factor. The changes in sea ice are analyzed based on the changes in the power ratio factor value, thereby effectively avoiding the high cost and low time resolution of sea ice detection in traditional methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 A flowchart of sea ice detection based on GNSS-IR power ratio observation according to an embodiment of the present invention; Figure 2 is a signal vector relationship diagram according to an embodiment of the invention; Figure 3 This is a time variation diagram of the daily average power ratio factor of the TUKT station S1 observation value in 2018 according to a specific embodiment of the present invention; Figure 4 This is a time variation diagram of the daily average power ratio factor of the TUKT station S2 observation value in 2018 according to a specific embodiment of the present invention; Figure 5 This is a time variation diagram of the daily average power ratio factor of the TUKT station S5 observation value in 2018 according to a specific embodiment of the present invention. DETAILED DESCRIPTION

[0011] Embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be understood as limiting the present invention.

[0012] Before introducing the sea ice detection method using GNSS-IR power ratio data, a brief introduction to the GNSS-IR technology used in sea ice detection is given.

[0013] Microwave remote sensing can achieve large-area, non-contact, real-time and continuous detection. Among them, GNSS-R technology, as a new application branch of GNSS, has developed rapidly in recent years with the continuous development and improvement of the global satellite navigation system.

[0014] With the development of satellite navigation technology, GNSS can not only be used for positioning, navigation and timing; in recent years, its application in remote sensing such as earthquake, meteorology, ocean, and land has also achieved breakthroughs and development. As an emerging passive remote sensing technology, GNSS-IR technology has the advantages of high accuracy, wide coverage, portable equipment, and no need for a specific signal source. Compared with GNSS-R, GNSS-IR technology does not require special receiver equipment, but only requires the right-hand circularly polarized navigation antenna used by the surveying and mapping receiver. In traditional satellite signal reception, in addition to the GNSS direct signal transmitted by line of sight, the receiver can also receive signals reflected by multiple paths at the same time, which can also be called multipath reflection signals. Although multipath signals are the source of errors in navigation and positioning, they have a certain correlation with the characteristics of the reflecting surface (such as the presence or absence of sea ice). For shore-based GNSS receivers, the direct signal and the reflected signal received by the receiver have almost the same frequency, but due to different propagation paths, the reflected signal has a path delay relative to the direct signal. Because GNSS-IR only needs to use the same antenna to receive two signals, a stable interference phenomenon is generated to form an interference signal.

[0015] GNSS-IR sea ice detection technology has the following advantages: Low cost: The completely free L-band signal transmitted by the navigation satellite is used as the signal source. The electromagnetic waves in this band are very sensitive to the changes in the electromagnetic characteristics of the sea ice surface, and the atmospheric attenuation is small. Due to the difference in the dielectric constants of liquid water and solid ice, the sea ice can be better monitored. In addition, the GNSS-IR technology adopts a heterogeneous observation mode, with separate transmitters and receivers. There is no need to design a separate transmitter, and the complexity and cost of the receiver are low, which reduces the size and weight of the equipment and makes it easy to build a ground-based platform.

[0016] A large number of signal sources: There are currently more than 32 GPS satellites available, and more signal sources will be provided in the future, including GLONASS satellites (rated at 24), China's BeiDou satellites (rated at 35), plus redundant satellites of each system and other regional navigation augmentation systems, the total number of satellites will exceed 150.

[0017] Realize all-weather detection: GNSS-IR uses L-band as microwave signal resources, which has the advantages of global coverage, all-weather, and high stability. It has a wide range of applications and is less affected by clouds, weather, and day and night.

[0018] Realize full-time detection: The design of GNSS in the space constellation makes it possible to receive at least four satellite signals at most places on the earth at any time. For the same location, continuous observation can be achieved at all times.

[0019] Based on the above reasons, an embodiment of the present invention proposes a sea ice detection method based on GNSS-IR power ratio data.

[0020] The following describes a sea ice detection method based on GNSS-IR power ratio data according to an embodiment of the present invention with reference to the accompanying drawings.

[0021] like Figure 1 As shown, the sea ice detection method based on GNSS-IR power ratio data includes the following steps: S101, obtain station observation data, SP3 files and measured sea ice comparison data, where the target data includes GNSS station observation data and SP3 files, as well as measured sea ice density comparison data downloaded from the ECWMF website. The station selected in this example is the TUKT station in northern Canada (69.44°N, 132.99°W). To ensure that all data comes from the sea surface, the selected altitude angle range is 5°~20°, and the azimuth angle range is 8°~125° (this range is the sea ice detection range). The station data can be directly downloaded from the official website and its compressed file can be batch converted to O file format.

[0022] It can be understood that the embodiment of the present invention first acquires data, and can directly use the compressed RINEX format data downloaded from the GNSS-IR website to acquire the required information from the O file of the observation data.

[0023] Specifically, the observation file O file of the corresponding measuring station is obtained from the GNSS-IR data download website, the SP3 file of the year corresponding to the observation file (for example, 2018 is selected in the example of the present invention) is obtained, and the sea ice density data of the corresponding year is downloaded from the ECWMF website and the sea ice change curve obtained by processing is used as the measured sea ice comparison data.

[0024] S102, obtaining low elevation angle data satisfying preset conditions according to the target data.

[0025] It can be understood that the reflection coefficient of the embodiment of the present invention decreases with the increase of the satellite altitude angle. In order to ensure that all the data processed by the calculation results come from the ocean reflection surface, the present invention extracts data within the range of altitude angle 5°~20° and azimuth angle 8°~125°.

[0026] Specifically, the embodiment of the present invention processes the above O file and the SP3 file downloaded from the website, and extracts the elevation angle, azimuth angle, UTC time and SNR data of different bands (for example, S1, S2 and S5 of the embodiment of the present invention) of each satellite in the GPS system from the observation file. The extracted data is screened to select the signal-to-noise ratio (SNR) data of different bands within a low elevation angle (for example, 5° to 25° in the embodiment of the present invention).

[0027] S103, obtaining the main frequency according to the spectrum analysis, calculating the effective antenna height, and thereby selecting data suitable for calculation.

[0028] It can be understood that the embodiment of the present invention sinusoids the elevation angle according to the low elevation angle data, and performs a third-order polynomial fitting on the SNR data to remove the direct component, obtain the SNR multipath component, and obtain the main frequency through signal spectrum analysis to calculate the effective antenna height.

[0029] Specifically, the signal-to-noise ratio (SNR) of the direct and reflected signals can be expressed as: ; in, , represent the direct signal amplitude and the reflected amplitude respectively; It represents the phase difference of the reflected signal relative to the direct signal. is the noise power.

[0030] In one embodiment of the invention, the main frequency is obtained by signal spectrum analysis, and the effective antenna height is calculated, which further includes: performing spectrum analysis through Lomb-Scargle transformation to obtain the spectrum of the frequency; calculating the effective antenna height according to the formula, and screening the effective data. The above data is divided into rising and falling sections according to the altitude angle, and converted into an altitude angle sinusoidal sequence, and finally a third-order polynomial fitting is performed to obtain the direct signal power, and the direct component is removed.

[0031] S104, using the third-order polynomial fitting data as the SNR direct component, that is, the direct signal power sequence.

[0032] Figure 2 is the signal vector relationship diagram. The composite signal for the direct and reflected signals, and are the direct signal and reflected signal components received by the GNSS antenna, is the angle between the reflected signal vector and the synthetic signal vector, is the angle between the direct signal vector and the reflected signal vector, is the phase of the direct signal, is the phase of the composite signal, is the phase difference between the direct signal and the composite signal; It is understandable that if Figure 2 As shown, the composite signal after extracting the multipath components It can be expressed as: ; In the formula, and are the direct signal and reflected signal components received by the GNSS antenna, represents the amplitude of the signal, is the phase noise. Among them, , , is the antenna gain with the satellite elevation angle as a variable, where the horizontal plane passing through the antenna phase center is defined as , below is , is the Fresnel reflection coefficient. Since the reflected signal may be lost or reflected multiple times during propagation, the direct signal power is much greater than the reflected signal power, that is: , the first two terms on the right side of the equation can be directly regarded as , so the signal-to-noise ratio (SNR) of the synthetic signal can be simplified to: ; In the formula, is the direct signal power component obtained by extracting the signal-to-noise ratio observation value from the RINEX observation file and fitting it with a third-order polynomial. It is the interference observation value generated by the direct signal and the reflected signal.

[0033] S105, obtaining a reflected signal power sequence through Hilbert transform, and obtaining a reflected / direct power ratio.

[0034] It can be understood that the Hilbert transform is used to process the residual signal-to-noise ratio sequence, and the transformed signal can be expressed as: ; in represents the residual signal-to-noise ratio sequence, is the Hilbert transform symbol, is the integration variable, , is the satellite altitude angle.

[0035] In order to make processing more convenient or meaningful, it is usually necessary to construct an analytical signal associated with the time domain real signal : ; The real part is the original reflected signal, is the imaginary number symbol, the imaginary part is the signal after Hilbert transformation; , is the satellite elevation angle. Since the essence of Hilbert transform is equivalent to an orthogonal filter, we can finally get the following by combining the Euler formula: ; in , is the vertical distance from the antenna phase center to the reflecting surface, is the signal wavelength, , is the satellite altitude angle, is the load wave signal, is the complex envelope. For the analytical signal The absolute value can be obtained to obtain the envelope of the analytical signal, that is, the amplitude : ; The amplitude sequence thus obtained is taken as the reflected signal power The direct signal power sequence is obtained by fitting the third-order polynomial, and the reflected signal power sequence is extracted by Hilbert transform. Therefore, the reflected / direct signal power ratio sequence It can be expressed as: ; S106, after removing invalid data, taking the average value of the remaining results according to different frequency points of different satellites as the daily average power ratio factor value, and statistically analyzing the time variation series of the power ratio factor values ​​of the corresponding years to detect sea ice changes.

[0036] It is understandable that a large amount of invalid data exists in the result calculated by step S105. Specifically, this part of the calculation results does not conform to the valid data range analyzed in step S102 and needs to be initially eliminated.

[0037] Furthermore, the average value of all satellite results in each day is calculated based on different frequency points as the time variation sequence of the power ratio factor value for detecting sea ice. Figure 3-5 As shown, Figure 3 is the time variation series of the daily average power ratio factor observed at TUKT station S1 in 2018, Figure 4 is the time variation series of the daily average power ratio factor observed at TUKT station S2 in 2018, Figure 5 This is the time variation series of the daily average power ratio factor observed at TUKT station S5 in 2018. Figure 3-5 The black line in the figure represents the sea ice density comparison data of the TUKT station, and the dotted data represents the daily average power ratio factor of the TUKT station obtained by the method of the present invention.

[0038] By comparing the sea ice density data of the TUKT station in 2018, it can be found that: when the sea ice melted on July 1, 2018, the power ratio factor value showed a significant downward trend, which shows that the reflecting surface changed from sea ice to sea water, and the increase in roughness led to a decrease in the factor value; when sea ice appeared on September 30, 2018, the power ratio factor value showed a significant upward trend, which shows that the reflecting surface changed from sea water to sea ice, and the decrease in roughness led to an increase in the factor value.

[0039] The results show that the variation trend of the daily average power ratio factor obtained by the method of the present invention can detect the appearance and melting of sea ice, and shows good temporal consistency compared with the sea ice density data of the observation station.

[0040] According to the sea ice detection method based on GNSS-IR power ratio data proposed in an embodiment of the present invention, the ratio of the reflected component power to the direct component power is obtained through the interference relationship between the satellite reflected signal and the direct signal, thereby detecting the appearance and melting time of sea ice, reducing the high cost of traditional inversion methods and improving the timeliness of monitoring.

[0041] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A sea ice detection method based on GNSS-IR power ratio data, characterized in that: include: Through the vector relationship between the direct signal, reflected signal and interference signal, the direct power sequence and the interference signal sequence are extracted from the signal-to-noise ratio SNR, and the interference signal is Hilbert transformed to obtain the reflected signal power sequence. Then, the reflected / direct signal power ratio sequence is obtained and averaged according to different frequencies of different satellites as the daily average power ratio factor, so as to use the time variation sequence of the power ratio to detect the appearance and melting of sea ice.

2. The sea ice detection method based on GNSS-IR power ratio data according to claim 1, characterized in that: Through the vector relationship between the direct signal, the reflected signal and the interference signal, the direct power sequence and the interference signal sequence are extracted from the signal-to-noise ratio SNR, specifically: When the reflected signal and the direct signal are received by the same antenna at the same time, an interference effect will occur. The power of this interference effect is measured by the signal-to-noise ratio recorded by the receiver. The mathematical expression of the signal-to-noise ratio SNR is: ； in, , represent the direct signal amplitude and the reflected amplitude respectively; It represents the phase difference of the reflected signal relative to the direct signal. is the noise power; The phase delay of the reflected signal compared to the direct signal It is expressed as: ； in: is the satellite altitude angle, is the signal wavelength; is the vertical distance from the antenna phase center to the reflecting surface; The trend term in the signal-to-noise ratio (SNR) is obtained by using a third-order polynomial fitting method. , which is the direct signal power; use the signal-to-noise ratio SNR to subtract the fitted direct signal component , thereby separating the interference signal component, that is, the residual signal-to-noise ratio dSNR; and because the geometric characteristics of the reflecting surface affect the roughness of the reflecting surface , which in turn causes the reflected signal to attenuate as the altitude angle changes. , so the attenuation It is expressed as: ； in is the signal wavelength, is the satellite altitude angle; , is the roughness of the reflective surface, is the RMS height of the reflecting surface; So the residual signal-to-noise ratio dSNR can finally be expressed as: ； in , is the satellite altitude angle, is the vertical distance from the antenna phase center to the reflecting surface, is the signal wavelength, is the residual phase of the reflected signal and the direct signal; , is the roughness of the reflective surface, is the root mean square of the height of the reflecting surface; the residual signal-to-noise ratio dSNR obtained is the interference signal sequence.

3. The sea ice detection method based on GNSS-IR power ratio data according to claim 1, characterized in that: The interference signal is Hilbert transformed to obtain the reflected signal power sequence, which is: The residual signal-to-noise ratio sequence is processed using the Hilbert transform. It is expressed as: ； in represents the residual signal-to-noise ratio sequence, is the Hilbert transform symbol, is the integration variable, , is the satellite altitude angle; Construct an analytic signal related to the real signal in the time domain : ； The real part is the original reflected signal, is the imaginary number symbol, the imaginary part is the signal after Hilbert transformation; , is the satellite altitude angle, and the analytical signal is finally obtained by combining the Euler formula for: ； in , is the vertical distance from the antenna phase center to the reflecting surface, is the signal wavelength, , is the satellite altitude angle, is the load wave signal, is the complex envelope; for the analytical signal The absolute value can be obtained to obtain the envelope of the analytical signal, that is, the amplitude : ； in is the original reflected signal, is the signal after Hilbert transformation; , is the satellite elevation angle, and the amplitude sequence obtained is used as the reflected signal power .

4. The sea ice detection method based on GNSS-IR power ratio data according to claim 1, characterized in that: The reflected / direct signal power ratio sequence is obtained and averaged according to different frequency points of different satellites as the daily average power ratio factor, so as to detect the appearance and melting of sea ice using the time variation sequence of the power ratio, specifically: Power ratio sequence of reflected signal to direct signal It is expressed as: ； in, is the reflected signal power, is the direct signal power; According to the frequency of different frequency points, the reflected / direct power ratio of satellite signals of the same frequency and different systems is averaged every day as the daily average power ratio factor. Long-term observation forms a time series, so as to detect the appearance and melting of sea ice according to the changes in the value of the time series factor.

Images