A method for high-precision real-time measurement of doppler frequency of deep space spacecraft

Abstract

This invention proposes a high-precision real-time method for measuring the Doppler frequency of deep-space spacecraft. The invention includes a reference signal construction module, a local correlation module, a phase analysis module, and a frequency calculation module. The Doppler frequency can be measured using only the downlink point frequency signal of the spacecraft. In use, this method first constructs a reference signal with a frequency close to the actual spacecraft Doppler signal received by each station, based on the high-precision atomic clock frequency standard of the station. Local correlation processing is then performed on the two signals to obtain the phase difference between the reference signal and the actual Doppler signal. This phase difference is then analyzed to obtain the frequency deviation between the reference signal and the actual Doppler signal. After correcting the reference frequency with this deviation value, the final Doppler frequency to be measured is obtained. For the Mars Express beacon at approximately 8.4 GHz, about 360 million kilometers from Earth, the real-time Doppler accuracy reaches mHz (a 1-second time interval), with a relative measurement accuracy of 10. -13 .

Description

Technical Field

[0001] This invention relates to the field of spacecraft orbit measurement and control technology, specifically to a method for measuring the Doppler frequency of deep spacecraft, which features passive real-time measurement of high-precision weak signal Doppler frequencies. Background Technology

[0002] High-precision Doppler frequency measurement is one of the fundamental methods for deep space radio orbit determination and navigation internationally. In 2001, the Jet Propulsion Laboratory (JPL) used two-way, three-way, and one-way Doppler data, including S-band and X-band data, to calculate the orbit of the lunar gravity field model MGS75D using MGS, Mariner9, Viking1, and Viking2. The accuracy of its S-band Doppler data velocity was approximately 1 mm / s (1-minute smoothing) and 0.3 mm / s (10-minute smoothing); the accuracy of X-band three-way and two-way Doppler data was approximately 0.05 mm / s (10-s integral), and the accuracy of X-band one-way Doppler data was approximately 1 mm / s (10-s integral). JPL also used two-way Doppler data when calculating the orbit of the MGS Mapping Phase, achieving a residual RMS better than 0.1 mm / s, a radial overlap accuracy better than 1 m, a track-direction overlap of approximately 10 m, and a normal overlap of approximately 2-4 m.

[0003] Traditional Doppler frequency measurement methods use the Doppler integration method, which mixes the frequency received by the ground receiver with the ground receiver's local oscillator frequency. Between two adjacent moments in the spacecraft signal, the number of beat cycles (i.e., Doppler count) is measured, and the spacecraft's Doppler frequency is deduced from the Doppler count. A drawback of this method is that if a cycle skip occurs, it directly affects subsequent counts, causing measurement errors. USB velocity and distance measurement is an active measurement method. Its principle is that the station transmits a signal to the spacecraft, which has a transponder. After receiving the signal, the spacecraft locks its frequency and transmits it back to the station. GPS satellites are geodetic satellites, and the signals they transmit carry on-board time codes. The station obtains the phase and frequency by measuring the code pseudorange or phase pseudorange.

[0004] The China National Space Administration (CNSA) and the Russian Federal Space Agency (Roscosmos) will jointly explore Mars, with China's spacecraft being the "Yinghuo-1" (YH-1). Due to limitations in weight and power consumption, the YH-1 spacecraft lacks a conventional telemetry and control transponder, instead employing only a beacon equipped with an ultra-stable crystal oscillator as a telemetry and control beacon. Consequently, the telemetry and control system cannot perform conventional two-way velocity and distance measurements on the YH-1. To determine the orbit of the YH-1, ground control can only employ a one-way Doppler frequency measurement method that only receives signals and does not transmit. This represents a new requirement for spacecraft measurement technology.

[0005] This invention addresses the need for unidirectional Doppler frequency measurement of spacecraft in my country's YH-1 Mars exploration project and the National 863 Major Project "Key Technologies for Milliarcsecond Precision VLBI Positioning and Orbit Determination." It proposes a passive method for measuring the Doppler frequency of spacecraft, which only requires unidirectional reception of the spacecraft's downlink point frequency signal to obtain the Doppler measurement frequency. Its advantages include no need for uplink signals from a station, system simplicity, good concealment, applicability to deep space, high accuracy, real-time signal acquisition, and simplified onboard beacon equipment for the target satellite. A search of relevant literature has revealed no similar technologies currently available. Summary of the Invention

[0006] The problem this invention aims to solve is to provide high-precision Doppler frequency data in real time for orbit determination of deep spacecraft.

[0007] The technical solution adopted by this invention to solve its technical problem is: to provide a method for passive real-time measurement of high-precision Doppler frequency of spacecraft. This method realizes the output of Doppler frequency, time scale, and frequency form error through a reference signal construction module, a local correlation module, a phase analysis module, and a frequency calculation module.

[0008] The reference signal construction module, based on the high-precision atomic clock frequency standard of the stations, constructs a reference signal with a frequency close to the actual spacecraft Doppler signal received by each station. This is implemented in three steps. Step 1: Analyze the spectrum using FFT to calculate the frequency ω0 of the signal to be determined, constructing a single-frequency reference signal. t is synchronized with the time of the observation data to be determined. The resolution of the spectrum and the local correlation step size satisfy 2|Δf|T. l The relationship is <1. Step 2: When the frequency form error after correlation does not meet the accuracy requirements, reconstruct the reference signal using the frequency obtained from the correlation. Step 3: To improve calculation speed or measurement accuracy, accurately measure the frequency at second t0 using the construction method in step 1 and the following steps, and use it as [t0-T]. b ,t0+T a [t0, t0+T] a [] or [t0-T] b ,t0](T a T b The reference frequencies (all greater than 0) are at a frequency change rate of approximately Under the premise:

[0009] The frequency at second t0 is used as [t0-T] b ,t0+T a When the reference frequency is [ ], it satisfies and

[0010] The frequency at second t0 is defined as [t0, t0+T]a When the reference frequency is [ ], it satisfies

[0011] The frequency at second t0 is used as [t0-T] b When the reference frequency is [t0], it satisfies

[0012] The reference signal construction module generates the reference signal for use by relevant local modules.

[0013] The local correlation module correlates the reference signal with the decoded observation data received by the station according to corresponding time segments, with each time segment being T. l For a sampling rate of T s The observation data, T l shared within Take 10 data points, sum them, filter out high-frequency terms, and calculate the phase of the difference-frequency term. Assume the total time of the data involved in the correlation is T. all Then a total of Each phase. The local correlation module outputs the phase difference between the reference signal and the signal received at the station, which is then used by the phase analysis module.

[0014] Within the phase analysis module, the ambiguity of the phase difference obtained from local correlation is first resolved to ensure continuous phase. It is assumed that the phase of the signal received by the station and the phase of the reference signal satisfy... The model (if the reference signal is) but If the reference signal is but The phase difference between the two is calculated. The phase obtained from local correlation is cyclically adjusted to remove outliers before fitting the model. The resulting fitting coefficients are compared with 'a' in the model. i One-to-one correspondence. The phase analysis module obtains the polynomial fitting relationship between the phase difference between the phase of the received signal at the station and the phase of the reference signal, which is then used by the frequency calculation module.

[0015] The frequency calculation module differentiates the phase difference polynomial fitting equation obtained by the phase analysis module to obtain the instantaneous frequency difference between the received signal and the reference signal at time t0. This allows us to obtain the frequency of the signal received by the measuring station. The formal error of the frequency is then obtained. When the formal error does not meet the accuracy requirements, an iterative correlation calculation method is used to reconstruct the reference signal using the frequency obtained from the previous correlation, and then re-correlate to obtain the Doppler frequency. If the number of iterations exceeds the maximum allowed number of iterations, the frequency with the smallest formal error is selected as the final result.

[0016] This invention enables the passive measurement of Doppler frequencies of spacecraft without relying on prior frequency information. It offers excellent concealment and high precision. For signals with a spectral density signal-to-noise ratio of 25dB, the measurement accuracy reaches 0.3mm / s, and the measurement distance is long (up to 360 million kilometers). For observation data with a sampling rate of 4Mbps and a bandwidth of 2MHz, the real-time performance is 1s. Attached Figure Description

[0017] Figure 1 This is a flowchart of the calculation process of the present invention.

[0018] Figure 2 This is a spectrum diagram of the spacecraft (Mars Express) signal.

[0019] Figure 3 See attached figure for Example 1

[0020] Figure 4 See attached figure for Example 2. Detailed Implementation

[0021] Appendix Figure 1 This is a flowchart illustrating a method for high-precision real-time measurement of the Doppler frequency of deep-space spacecraft. As shown in the diagram, this invention consists of a reference signal construction module, a local correlation module, a phase analysis module, and a frequency calculation module. Its working principle is as follows:

[0022] Assume the spacecraft is at t i The frequency of transmission at any given time is ω i After Δτ ij The time period is transmitted to the station. The station is at t j (t j =t i +Δτ ij The frequency of the carrier signal received at time ω j The sampling time interval at each station is T. s The amplitude of the signal acquired by the station terminal is A(t), the phase of the received signal is Φ(t), and the noise is N(t), all following a Gaussian distribution with a mean of 0. Then, at any time t (t = nT)... s (where n is any integer), the signal received by the station can be expressed as

[0023] P(t)=A(t)exp[-jΦ(t)]+N(t) (1)

[0024] By using FFT to analyze the spectrum of the signal received by the station, the coarse frequency ω0 of the signal to be determined is obtained, and a reference signal R(t) = exp(jω0t) is constructed. The FFT resolution is Δf.

[0025] Correlate the reference signal with the signal received at the station. The time is the starting point of time, time 0, where T allLet T be the total relevant time length. Then, in [0, T]... all Within [ ], the phase of the reference signal and Φ(t) satisfy the following relationship:

[0026]

[0027] in, For the phase difference, a i For the relationship coefficient. Considering the characteristics of the received signal, the phase difference may satisfy a first-order linear relationship, a second-order, third-order, sixth-order, or higher-order functional relationship during actual processing. The principle description of this invention only uses a second-order functional relationship as an example. Other higher-order functional relationships are also within the scope of protection of this patent. Equation (1) can be described as

[0028] P(t)=A(t)exp[-j(ω0t-a0-a1t-a2t 2 )]+N(t) (3)

[0029] If the received signal is correlated with the reference signal, then

[0030]

[0031] In the relevant process, in order to improve the signal-to-noise ratio, T all Integrate the data piecewise over a time period. Assume the step size for piecewise integration is T. l To ensure that the signal is not distorted during correlation, 2|Δf|T must be satisfied. l <1.

[0032] Any T l During the time period, equation (4) is expressed as

[0033]

[0034] The time interval t is the sampling time T. s The value is typically on the order of microseconds, or varies depending on the actual sampling frequency. The above summation can be converted into integration. The phase in equation (5) can be approximated using the rectangular integration method.

[0035]

[0036] Where N takes values ​​ranging from 1 to 10. T all A series of phase deblurring results are obtained within a time period and fitted. Outliers are then iteratively removed according to the 3-sigma rule, and the formal error is calculated. The fitting coefficients a0, a1, and a2 are obtained.

[0037]

[0038] ω j =ω(t) j )

[0039]

[0040] σ ω For t j The formal error of the time-frequency. When the formal error does not meet the measurement accuracy requirements, the obtained ω is... j Construct a reference signal and recalculate iteratively.

[0041] When the signal received by the station is expressed as P(t)=A(t)exp[jΦ(t)]+N(t), then the constructed reference signal is R(t)=exp(-jω0t), and the obtained Doppler frequency is

[0042] When measuring the Doppler frequency of a spacecraft, to improve the calculation speed, the reference frequency and T can be flexibly selected based on the characteristics of the spacecraft's Doppler frequency variation. l and T all Assume T all If the difference between the frequency at any given moment and the reference frequency is Δf, then 2|Δf|T satisfies the condition. l The relationship is <1; assuming the rate of change of frequency is... T all The time difference ΔT between any instant and the reference frequency within the range satisfies the following condition. The relationship.

[0043] Example

[0044] Mars Express is a European Space Agency satellite that observes Mars, approximately 360 million kilometers from Earth. (Attached) Figure 2 This is a spectrum of the X-band signal from a certain station observing the Mars Express rover, with a sampling rate of 16 Mbps, a bandwidth of 8 MHz, and 1-bit sampling. Using data from 2 hours 17 minutes UTC (02h17m) as an example, the Doppler frequency was calculated using Matlab software on a high-performance computer. The calculation was limited to a maximum frequency error of 5 mHz and a maximum of 3 iterations.

[0045] Example 1

[0046] 1) Purpose: such as Figure 2 The frequency of the downlink primary carrier of the Mars Express X-band beacon was measured, and its signal-to-noise ratio was approximately 27 dB. The frequency of UTC02h17m05.5s was calculated.

[0047] 2) Implementation steps:

[0048] Reference signal construction: The initial frequency of the main carrier is calculated using FFT with a resolution of approximately 30Hz. The approximate frequency of the main carrier is 3214904.7852Hz. The reference signal exp(j2π*3214904.7852t) is constructed and synchronized with the hydrogen atom clock.

[0049] Local correlation: The reference signal is locally correlated with the data from the 5th to 6th second, with a correlation step size of 0.01 seconds, to obtain the phase difference ( Figure 3 (See above image);

[0050] Phase analysis: After resolving the ambiguity of the phase difference, a quadratic fitting is performed to obtain the polynomial parametric relationship of the phase difference as ΔΦ = 4.4241t. 2 +0.33469t-0.43529, in weeks ( Figure 3 (Middle and lower image);

[0051] Frequency calculation: ΔΦ = 4.4241t 2 Differentiating +0.33469t - 0.43529, we obtain that the reference frequency differs from the actual Doppler frequency at 5.5s by 4.7588Hz. Therefore, the frequency at 02h17m05.5s is 3214900.0264Hz, with a formal error of 2.6mHz, satisfying the aforementioned requirements. From ΔΦ = 4.4241t... 2 The relationship +0.33469t-0.43529 can be used to obtain the frequency at any time within 02h17m5-6s.

[0052] Example 2

[0053] 1) Objective: To measure the frequency of the downlink sidetone of the Mars Express X-band beacon, which has a signal-to-noise ratio of approximately 11 dB, about 15 dB lower than the main carrier (see...). Figure 2 ), calculate the frequency of UTC02h17m05.5s.

[0054] 2) Implementation steps:

[0055] Reference signal construction: The frequency of 2166279.412Hz at 5.5s was obtained using the method described in Example 1 (formal error 8mHz, after 3 iterations). A single-frequency reference signal exp(j2π*2166279.412t) was constructed, and the reference signal was synchronized with the hydrogen atom clock.

[0056] Local correlation: The single-frequency reference signal is correlated with data from 0s to 10s, with a correlation step size of 0.01s, to obtain the phase difference ( Figure 4 (Middle and upper image);

[0057] Phase analysis: The ambiguity is resolved by calculating the phase difference obtained from local correlation, and a second fitting is performed to remove outliers, resulting in the polynomial parametric relationship of the phase difference as ΔΦ = 4.4209t. 2-48.6197t + 15.0078, in weeks ( Figure 4 (Middle and lower image);

[0058] Frequency calculation: ΔΦ = 4.4209t 2 Differentiating -48.6197t + 15.0078, we obtain that the Doppler frequency of the reference signal differs from that of the 5.5s Doppler frequency by 0.0102Hz, i.e., 02h17m05.5s is 2166279.4018Hz, with a formal error of 0.4mHz; from ΔΦ = 4.4209t 2 The relation -48.6197t+15.0078 can be used to obtain the frequency at any time within 02h17m0-10s.

Claims

1. A method for high-precision real-time measurement of the Doppler frequency of a deep-space spacecraft, wherein the method, without relying on prior frequency information, passively and in real-time measures the Doppler frequency of spacecraft signals, characterized in that... It consists of a reference signal construction module, a local correlation module, a phase analysis module, and a frequency calculation module, which realizes the output of Doppler frequency, time scale, and frequency form error. The reference signal construction module constructs a reference signal with a frequency close to the spacecraft Doppler signal received by each station based on the high-precision atomic clock frequency standard of the station. The constructed reference signal is a single-frequency signal, which tracks the change of the actual Doppler signal frequency in real time, so that the frequency of the reference signal is close to the frequency of the Doppler signal received by the station. The constructed reference signal is synchronized with the signal received by the station. The construction of the reference signal involves three steps: Step 1: Analyze the spectrum of the received signal from the station using FFT to calculate the frequency ω0 of the signal to be determined, and construct a single-frequency reference signal. The resolution Δf of the spectrum and the local correlation step size T l Satisfy 2|Δf|T l The relationship is <1; Step 2: After the reference signal is correlated with the signal received by the station, if the frequency form error does not meet the accuracy requirements, the reference signal is reconstructed using the frequency obtained from the correlation. Step 3: To improve calculation speed or measurement accuracy, the method constructed in step 1 and the frequency of accurately measuring the t0th second are used as [t0-T]. b ,t0+T a [t0, t0+T] a [] or [t0-T] b The reference frequency within [t0] is such that the frequency changes at a rate of t0. Under the premise: The frequency at second t0 is used as [t0-T] b ,t0+T a When the reference frequency is [ ], it satisfies and The frequency at second t0 is defined as [t0, t0+T] a When the reference frequency is [ ], it satisfies The frequency at second t0 is used as [t0-T] b When the reference frequency is [t0], it satisfies 2. The method for high-precision real-time measurement of Doppler frequency of deep spacecraft according to claim 1, characterized in that, The local correlation module performs segmented correlation processing on the constructed reference signals of each station and the corresponding received signals of the stations, with a time length of T. l The correlation results are summed, high-frequency terms are filtered out, and the phase of the difference frequency term is calculated. When the difference between the reference signal frequency and the true signal frequency is Δf, the correlation time length is... Obtain the phase difference between the reference signal and the signal received by the station.

3. The method for high-precision real-time measurement of Doppler frequency of deep spacecraft according to claim 1, characterized in that, The phase analysis module uses a polynomial parametric model to describe the relationship between the spacecraft Doppler phase of the received signal and the phase of the reference signal: If the reference signal is but If the reference signal is but The phase difference between the reference signal and the actual Doppler signal is used; ambiguity is removed from the obtained phase difference, outliers are eliminated, and the least squares method is used to estimate 'a'. i The functional relationship between the phase difference between the reference signal and the actual Doppler signal is obtained.

4. The method for high-precision real-time measurement of Doppler frequency of deep spacecraft according to claim 1, characterized in that, The formula for calculating the frequency is as follows: Where t0 is the time to be determined, and a i for The estimated parameters are used to calculate ω using the least squares method. rec Formal error.

5. The method for high-precision real-time measurement of the Doppler frequency of deep spacecraft according to claim 4 adopts an iterative correlation calculation method: when the formal error does not meet the requirements, the previously calculated ω is used. rec The reference signal is reconstructed and iteratively correlated with the signal received by the station. The maximum number of iterations is set to m, where m > 0. If the accuracy requirement is still not met after m iterations, the measurement frequency with the smallest formal error is selected as the final result.

Images