A method for obtaining pulsar radiation signal profiles based on overlapping region folding
By employing a contour folding method with overlapping regions and a spectral truncation technique, the problem of low contour quality of X-ray pulsar radiation signals under short-term observation conditions was solved, achieving more efficient information utilization and improved signal contour quality.
Patent Information
- Application Number
- CN202311089818.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-28
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2043-08-28
AI Technical Summary
Under short-term observation conditions, existing technologies struggle to obtain high-quality X-ray pulsar radiation signal profiles, and the correlation coefficient between the signal profiles and standard profiles is low.
The contour folding method based on overlapping intervals is adopted. The photon TOA or phase sequence is divided into subsequences with overlapping intervals, and the number of photons or phases in each bin is folded into the first subsequence. Combined with fast Fourier transform and spectral truncation, the signal contour is obtained.
It improved the correlation coefficient between the signal profile and the standard profile, increased the information utilization rate of the observation data, and improved the quality of the signal profile.
Smart Images

Figure CN117191006B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of signal processing technology and relates to a method for obtaining the contour of X-ray pulsar radiation signals, specifically a method for obtaining the contour of pulsar radiation signals based on overlapping interval folding, which can be used for processing astronomical data. Background Technology
[0002] X-ray pulsars are pulsars that emit X-ray signals. They rotate extremely rapidly, with frequencies reaching hundreds of hertz, and possess excellent rotational stability. When the radiation beam of an X-ray pulsar sweeps across a signal detector, the instrument records the stable and periodic X-ray pulsar radiation signal. However, because X-ray pulsars are extremely far from Earth, the radiation signal attenuates significantly during propagation. Therefore, the detector records the time of arrival (TOA) of each X-ray pulsar photon, a randomly distributed, non-equally spaced time sequence. Given these characteristics of the radiation signal, researchers hope to use this to construct navigation observations for autonomous spacecraft navigation, thus proposing X-ray pulsar navigation. The profile of the X-ray pulsar radiation signal describes the waveform characteristics of the signal. As important characteristic information of X-ray pulsars, it is crucial for obtaining observations for X-ray pulsar navigation. The quality of the signal profile is usually measured by the signal-to-noise ratio (SNR) or the correlation coefficient between the observed profile and the standard profile of the X-ray pulsar radiation signal. A higher SNR or correlation coefficient indicates better signal profile quality. In practical applications, given the limitations of pulsar radiation flux, the accuracy of existing detection devices, and the timeliness requirements of navigation missions, obtaining high-quality signal profiles under the condition of short-term observation with a time span of only a few hundred pulsar radiation signal cycles is of significant research value.
[0003] Currently, there are two basic methods for obtaining signal profiles from X-ray pulsar observation data.
[0004] The first method is periodic folding. This method divides the observed photon arrival time series into time periods, accumulates the photons into equally divided bins within a signal period, counts the number of photons in each bin, obtains the folding profile, and uses it as the signal profile.
[0005] The second method is epoch folding. This method first calculates the phase of each photon's arrival time relative to the reference epoch, then divides it according to the phase period, and then accumulates the photons into equally divided bins within a phase period. The number of photons in each bin is counted to obtain the folding profile, which is then used as the signal profile.
[0006] The performance of the above two methods is limited by the selection of the number of bins and the amount of observation data. Under short-term observation conditions, it is difficult to obtain high-quality signal profiles. Therefore, researchers have made improvements based on these methods. For example, the patent application CN110986922A, entitled "Method for Obtaining High Signal-to-Noise Ratio Profiles from Short-Term X-ray Pulsar Observations," first performs periodic or epochal folding on the observed photon arrival time (TOA) sequence to obtain the folded profile of the X-ray pulsar radiation signal. Then, it obtains the spectrum of the folded profile and extracts the spectral components with high signal-to-noise ratio from the spectrum, obtaining the signal profile through inverse Fourier transform. This method obtains the signal profile by acquiring the spectrum of the signal profile and performing an inverse Fourier transform on the spectrum, avoiding the negative impact of the selection of the number of bins on the signal profile quality during periodic or epochal folding, and improving the correlation coefficient between the signal profile and the standard profile. However, under short-term observation conditions, this method only uses the observation data once, resulting in low information utilization of the observation data. Therefore, the correlation coefficient between the obtained signal profile and the standard profile is low, and the quality of the signal profile still has room for improvement. Summary of the Invention
[0007] The purpose of this invention is to address the shortcomings of the prior art by proposing a method for obtaining pulsar radiation signal profiles based on overlapping interval folding, which solves the technical problem of low pulsar radiation signal profile quality under short-time observation conditions in the prior art.
[0008] To achieve the above objectives, the technical solution adopted by the present invention includes the following steps:
[0009] (1) Initialize parameters:
[0010] Initialization at observation time [t] a ,t b The TOA sequence of I X-ray pulsar photons detected by the spacecraft is as follows: The overlap ratio is d, the overlap value is w, and the period of the X-ray pulsar radiation signal is T. P , where t i Let I be the TOA time when the i-th X-ray pulsar photon arrives at the spacecraft, where I > 10;
[0011] (2) Obtaining the folded profile of X-ray pulsar radiation signal by using a contour folding method with overlapping regions:
[0012] (2a) Calculate the TOA time t for each photon based on the X-ray pulsar rotation phase model. i Corresponding phase The result contains I phase values phase sequence
[0013] (2b) Calculation of photon TOA sequence by overlap ratio d The number J of subsequences that can be divided is used to calculate the J time intervals on which the subsequences are based. And based on Photon TOA sequence Divide into J subsequences with overlapping intervals; or calculate the phase sequence using the overlap value w. The number K of the possible subsequences can be divided into, and the K phase intervals used to calculate the subsequences are calculated using K. And based on phase sequence Divide into K subsequences with overlapping intervals; where T j To divide the photon TOA sequence The time interval Φ is used to determine the j-th subsequence. k To divide the phase sequence The phase interval upon which the k-th subsequence is based;
[0014] (2c) The photon TOA time t in the J subsequences i All are reduced to [0,T] P The time interval is used to obtain the reduced photon TOA time t. ni and [0,T P The time interval is divided into M equal bins. The number of photons λ in the m-th bin of the j-th subsequence is counted. j (t m ); or the phases of K subsequences All phases are reduced to the phase interval [0,1) to obtain the reduced phase. The phase interval [0,1) is divided into M equal bins, and the number of photons in the m-th bin of the k-th subsequence is counted. Where M∈[2 11 (j=1,2,…,J,k=1,2,…,K,m=1,2,…,M);
[0015] (2d) Photon TOA sequence The number of photons λ in the m-th bin of the J subsequences j (t m Fold it into the first subsequence, or fold the phase sequence. The number of photons in the m-th bin block of the K subsequences Folding into the first subsequence yields the folded profile of the X-ray pulsar radiation signal containing M bins.
[0016] (3) Obtain the spectrum of the folded profile of the X-ray pulsar radiation signal:
[0017] Folded profile of X-ray pulsar radiation signal Performing an M-point Fast Fourier Transform yields a spectrum containing M spectral components.
[0018] (4) Obtain the spectrum of the denoised folded contour:
[0019] Using q as the spectral cutoff point, the spectrum of the folded contour The first q spectral components The truncation is performed and used as the spectrum of the denoised folded contour, where q is a positive integer;
[0020] (5) Obtain the contour of the X-ray pulsar radiation signal:
[0021] Spectrum of the denoised folded profile Perform an inverse fast Fourier transform to obtain the profile λ of the X-ray pulsar radiation signal.
[0022] Compared with the prior art, the present invention has the following advantages:
[0023] This invention obtains the folded profile of X-ray pulsar radiation signals by folding the number of photons in each bin of each photon TOA subsequence into the first subsequence, or by folding the number of photons in each bin of each phase subsequence into the first subsequence. This enables the reuse of observation data in overlapping regions of adjacent subsequences, increases the number of subsequences, and utilizes the information in the observation data more efficiently. It avoids the impact of periodic or epoch folding, which only uses the observation data once, on the utilization rate of observation data information in existing technologies, and improves the correlation coefficient between the signal profile and the standard profile, thereby improving the quality of the signal profile. Attached Figure Description
[0024] Figure 1 This is a flowchart illustrating the implementation of the present invention.
[0025] Figure 2 This is a schematic diagram illustrating the subsequence division of the X-ray pulsar photon TOA sequence into overlapping regions in this invention.
[0026] Figure 3 This is a comparison chart of the average correlation coefficients between the contours obtained from the processing of pulsar measured data by the present invention and existing technologies and the standard contours. Detailed Implementation
[0027] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0028] Reference Figure 1 The present invention includes the following steps:
[0029] (1) Initialize parameters:
[0030] Initialization at observation time [t] a ,t b The TOA sequence of I X-ray pulsar photons detected by the spacecraft is as follows: The overlap ratio is d, the overlap value is w, and the period of the X-ray pulsar radiation signal is T. P , where t i Let I be the TOA time when the i-th X-ray pulsar photon arrives at the spacecraft, where I > 10.
[0031] In this example, taking Crab pulsar as an example, the observation data of Crab pulsar by the US Rossi RXTE satellite is selected. The data packet number is 90802_02_02_00_01. The observation duration of this data packet is 913 seconds, and the number of photons observed is 9.46 million. The observation data of the first second is extracted as the photon arrival time series, and the value of I is 10371. The overlap ratio d and the overlap value w are both in the range of [0.5, 0.7], and the overlap value w is taken as 0.5.
[0032] (2) Obtaining the folded profile of X-ray pulsar radiation signal by using a contour folding method with overlapping regions:
[0033] In power spectrum estimation methods for stochastic processes, the Bartlett improved periodogram method divides the signal sampling sequence into several subsequences of equal length without overlap, calculates the power spectrum of each subsequence, and then sums and averages them to obtain a signal power spectrum with reduced spectral estimation variance. This method shares a similar underlying principle with existing techniques. The Welch improved periodogram method divides the signal sampling sequence into equal-length subsequences with overlap. When its subsequence length is the same as that of the Bartlett improved periodogram method, the number of its subsequences is greater than that of the Bartlett improved periodogram method. Therefore, the signal power spectrum estimation result obtained after summing and averaging the power spectra of each subsequence is better, and its spectral estimation variance is lower than that of the Bartlett improved periodogram method. Given this advantage of the Welch improved periodogram method, a contour folding method with overlapping intervals is designed to improve the quality of the signal contour.
[0034] For different application scenarios, researchers have designed periodic folding and epochal folding based on the same contour folding concept, using photon arrival time and phase respectively. Based on this characteristic, when designing contour folding methods with overlapping regions, photon arrival time and phase must also be used separately.
[0035] (2a) Calculate the TOA time t for each photon based on the X-ray pulsar rotation phase model. i Corresponding phase The result contains I phase values phase sequence
[0036] The X-ray pulsar rotation phase model describes the photon TOA time t i Its corresponding phase The relationship between these factors allows us to calculate the phase of any photon at time TOA using this model, which can be expressed in Taylor series expansion as follows:
[0037]
[0038] in Let f be the phase of the X-ray pulsar at reference time t0, and f be the rotation frequency of the X-ray pulsar. and These are the first and second derivatives of f, respectively.
[0039] (2b) Calculation of photon TOA sequence by overlap ratio d The number J of subsequences that can be divided is used to calculate the J time intervals on which the subsequences are based. And based on Photon TOA sequence Divide into J subsequences with overlapping intervals, and calculate them using the following formulas:
[0040]
[0041] T j :[t a +(j-1)(1-d)T P ,t a +T P +(j-1)(1-d)T P ), j=1,2,…,J
[0042] Where T j To divide the photon TOA sequence The time interval t used to determine the j-th subsequence a and t b These are the start and end times of the observation period, respectively. This is for rounding up;
[0043] Alternatively, the phase sequence can be calculated using the overlap value w. The number K of the possible subsequences can be divided into, and the K phase intervals used to calculate the subsequences are calculated using K. And based on phase sequence Divide the sequence into K subsequences with overlapping intervals, and calculate them using the following formulas:
[0044]
[0045]
[0046] Where Φ k To divide the phase sequence The phase interval used for the kth subsequence, and These represent the first and I-th photon phases, respectively. The process is as follows: Figure 2 As shown, photon data located in the overlapping region between two adjacent subsequences are reused, enabling more efficient use of the information contained in the observation data.
[0047] (2c) The photon TOA time t in the J subsequences i All are reduced to [0,T] P The time interval is used to obtain the reduced photon TOA time t. ni and [0,T P The time interval is divided into M equal bins. The number of photons λ in the m-th bin of the j-th subsequence is counted. j (t m The calculation formula is:
[0048]
[0049] in This is a floor function;
[0050] Or consider the phases of the K subsequences All phases are reduced to the phase interval [0,1) to obtain the reduced phase. The phase interval [0,1) is divided into M equal bins, and the number of photons in the m-th bin of the k-th subsequence is counted. The calculation formula is:
[0051]
[0052] Where M∈[2 11 (j=1,2,…,J), k=1,2,…,K, m=1,2,…,M.
[0053] (2d) Photon TOA sequence The number of photons λ in the m-th bin of the J subsequences j (t mFold it into the first subsequence, or fold the phase sequence. The number of photons in the m-th bin block of the K subsequences Folding into the first subsequence yields the folded profile of the X-ray pulsar radiation signal containing M bins.
[0054]
[0055]
[0056] in For folded outline The number of photons in the m-th bin.
[0057] This example uses, but is not limited to, phase sequences. Perform contour folding on overlapping regions with M bins, where M is 2. 20 The number of subsequences K is 59, the reference time t0 in the rotation phase model is 3.326400040037602e+08, and the initial phase... The value is 9.889587091314340e+09, the spin frequency f is 29.7925564233089, and the first derivative of the frequency is... The value is -3.73397D-10, the second derivative of the frequency. The value is 8.00D-21.
[0058] (3) Obtain the spectrum of the folded profile of the X-ray pulsar radiation signal:
[0059] Folded profile of X-ray pulsar radiation signal Performing an M-point Fast Fourier Transform yields a spectrum containing M spectral components.
[0060]
[0061] (4) Obtain the spectrum of the denoised folded contour:
[0062] As we know from signal processing, the spectrum of a signal is mainly distributed in the low frequency range, while the spectrum of noise is distributed across the entire frequency domain. The spectrum of X-ray pulsar radiation signals follows a power-law distribution, and its useful information is concentrated in the low-frequency part. High-frequency noise components can be removed by frequency domain truncation.
[0063] Using q as the spectral cutoff point, the spectrum of the folded contour The first q spectral components The truncation is performed and used as the spectrum of the denoised folded contour, where q is a positive integer.
[0064] In this example, the K_T method is used to calculate the spectral cutoff point, and the cost function is constructed using the mean square error of the integral (MISE).
[0065]
[0066] Where S m The spectrum containing the first m spectral components is the standard profile of the X-ray pulsar radiation signal. The spectrum of the folded contour containing the first m spectral components. and S m and conjugate, Let q be the cost function value. This cost function value exhibits a property of first monotonically decreasing and then monotonically increasing as the independent variable q increases. There is a minimum point. When the cost function value reaches the minimum, the corresponding q value is the optimal spectral cutoff point. Finally, the spectral cutoff point q is 6.
[0067] (5) Obtain the contour of the X-ray pulsar radiation signal:
[0068] Spectrum of the denoised folded profile A fast inverse Fourier transform is performed to obtain the contour λ of the X-ray pulsar radiation signal. The specific implementation process is as follows:
[0069] (5a) using the spectrum The first spectral component in DC component right Perform an inverse fast Fourier transform to calculate the DC component λ of the profile. D :
[0070]
[0071] (5b) Spectrum The first spectral component in Setting the value to 0 yields a spectrum consisting of q spectral components that do not contain a DC component. For the spectrum After performing an N-point inverse fast Fourier transform and taking its real part, a contour λ of length N without DC component is obtained. N :
[0072]
[0073]
[0074] in and Spectrum and spectrum The (m+1)th spectral component, where N is a positive integer, and Re is the operation for finding the real part of the complex number;
[0075] (5c) Through the profile λ that does not contain a DC component N DC component λ of the profile D Calculate the profile λ of the X-ray pulsar radiation signal:
[0076] λ=λ N +λ D .
[0077] In this example, the standard profile of the Crab pulsar radiation signal is selected as the reference profile, and the correlation coefficient between the reference profile and the profile obtained in this invention is calculated to be 0.8890.
[0078] The technical effects of this invention will be explained below through experiments based on satellite measurement data.
[0079] 1. Experimental conditions and contents:
[0080] This experiment uses observational data of Crab pulsar from the US Rossi RXTE satellite, data package number 90802_02_02_00_01. The computer parameters used in the experiment are: CPU i7-7700, RAM 16GB, operating system Windows 10, and software MATLAB R2021b.
[0081] The experiment involved using observed Crab pulsar data from the RXTE satellite. For 291 observation durations within a time range of 0.1s-3s, with intervals of 0.01s, 300 data segments were extracted from each observation duration. The overlap value w of the subsequence phase intervals was 0.5. Signal profiles were acquired using both the high signal-to-noise ratio profile acquisition method for short-term X-ray pulsar observations and the method of this invention. Using the standard profile of the Crab pulsar radiation signal as a reference profile, the average correlation coefficient between profiles at different observation durations was calculated. The results are shown below. Figure 3 As shown.
[0082] 2. Analysis of experimental results:
[0083] Reference Figure 3The horizontal axis represents the observation time of the satellite's measured data, with 291 observation duration nodes ranging from 0.1s to 3s. The vertical axis represents the average correlation coefficient between contours for each observation duration. As can be seen from the figure, the average correlation coefficient between contours obtained by both methods increases with the extension of the observation time. The average correlation coefficient between contours obtained by this invention consistently surpasses that of existing technologies. Furthermore, under short-term observation conditions, the average correlation coefficient between contours obtained by this invention is even higher, indicating that this invention can obtain higher-quality X-ray pulsar radiation signal contours.
[0084] The above description is merely a specific example of the present invention and does not constitute any limitation on the present invention. Obviously, those skilled in the art, after understanding the content and principles of the present invention, may make various modifications and changes in any form and detail without departing from the principles and structure of the present invention. However, these modifications and changes based on the ideas of the present invention are still within the scope of protection of the claims of the present invention.
Claims
1. A method for obtaining the contour of pulsar radiation signals based on overlapping interval folding, characterized in that, Includes the following steps: (1) Initialize parameters: Initialization at observation time [t] a ,t b The TOA sequence of I X-ray pulsar photons detected by the spacecraft is as follows: The overlap ratio is d, the overlap value is w, and the period of the X-ray pulsar radiation signal is T. P , where t i Let I be the TOA time when the i-th X-ray pulsar photon arrives at the spacecraft, where I > 10; (2) Obtaining the folded profile of X-ray pulsar radiation signal by using a contour folding method with overlapping regions: (2a) Calculate the TOA time t for each photon based on the X-ray pulsar rotation phase model. i Corresponding phase The result contains I phase values phase sequence (2b) Calculation of photon TOA sequence by overlap ratio d The number J of subsequences that can be divided is used to calculate the J time intervals on which the subsequences are based. And based on Photon TOA sequence Divide into J subsequences with overlapping intervals; or calculate the phase sequence using the overlap value w. The number K of the possible subsequences can be divided into, and the K phase intervals used to calculate the subsequences are calculated using K. And based on phase sequence Divide into K subsequences with overlapping intervals; where T j To divide the photon TOA sequence The time interval Φ is used to determine the j-th subsequence. k To divide the phase sequence The phase interval upon which the k-th subsequence is based; (2c) The photon TOA time t in the J subsequences i All are reduced to [0,T] P The time interval is used to obtain the reduced photon TOA time t. ni and [0,T P The time interval is divided into M equal bins. The number of photons λ in the m-th bin of the j-th subsequence is counted. j (t m ); or the phases of K subsequences All phases are reduced to the phase interval [0,1) to obtain the reduced phase. The phase interval [0,1) is divided into M equal bins, and the number of photons in the m-th bin of the k-th subsequence is counted. Where M∈[2 11 (j=1,2,…,J,k=1,2,…,K,m=1,2,…,M); (2d) Photon TOA sequence The number of photons λ in the m-th bin of the J subsequences j (t m Fold it into the first subsequence, or fold the phase sequence. The number of photons in the m-th bin block of the K subsequences Folding into the first subsequence yields the folded profile of the X-ray pulsar radiation signal containing M bins. (3) Obtain the spectrum of the folded profile of the X-ray pulsar radiation signal: Folded profile of X-ray pulsar radiation signal Performing an M-point Fast Fourier Transform yields a spectrum containing M spectral components. (4) Obtain the spectrum of the denoised folded contour: Using q as the spectral cutoff point, the spectrum of the folded contour The first q spectral components The truncation is performed and used as the spectrum of the denoised folded contour, where q is a positive integer; (5) Obtain the contour of the X-ray pulsar radiation signal: Spectrum of the denoised folded profile Perform an inverse fast Fourier transform to obtain the profile λ of the X-ray pulsar radiation signal.
2. The method according to claim 1, characterized in that, In step (2a), the TOA time t for each photon is calculated based on the X-ray pulsar rotation phase model. i Corresponding phase The calculation formula is: in Let f be the phase of the X-ray pulsar at reference time t0, and f be the rotation frequency of the X-ray pulsar. and These are the first and second derivatives of f, respectively.
3. The method according to claim 1, characterized in that, The photon TOA sequence described in step (2b) The number of subsequences J that can be divided into, and the division of the photon TOA sequence. The time interval T used for the j-th subsequence j Phase sequence The number of subsequences K into which it can be divided and the phase sequence of the division The phase interval Φ used for the k-th subsequence k The calculation formulas are as follows: T j :{t a +(j-1)(1-d)T P ,t a +T P +(j-1)(1-d)T P ),j=1,2,…,J Where t a and t b These are the start and end times of the observation period, respectively. and These are the first and I-th photon phases, respectively. This is for rounding up.
4. The method according to claim 1, characterized in that, Step (2c) describes setting the TOA time t of photons in the J sub-sequences. i All are reduced to [0,T] P The time interval and the phase of the K subsequences) All phase intervals are reduced to [0,1), and the calculation formulas are as follows: in For floor operations, t ni For the reduced photon TOA time, This is the phase after reduction.
5. The method according to claim 1, characterized in that, The photon TOA sequence described in step (2d) The number of photons λ in the m-th bin of the J subsequences j (t m Fold it into the first subsequence, or fold the phase sequence. The number of photons in the m-th bin block of the K subsequences The folding formulas for folding to the first subsequence are as follows: in For folded outline The number of photons in the m-th bin.
6. The method according to claim 1, characterized in that, The folding profile of the X-ray pulsar radiation signal described in step (3) The formula for performing the M-point Fast Fourier Transform is as follows:
7. The method according to claim 1, characterized in that, The formula for calculating the frequency spectrum cutoff point as mentioned in step (4) is as follows: Where S m The spectrum containing the first m spectral components is the standard profile of the X-ray pulsar radiation signal. The spectrum of the folded contour containing the first m spectral components. and S m and conjugate, The cost function value is q, and each q value corresponds to a cost function value. When the cost function value reaches its minimum, the corresponding q value is the optimal spectral cutoff point.
8. The method according to claim 1, characterized in that, The spectrum of the denoised folded profile described in step (5) The steps to perform an inverse fast Fourier transform are as follows: (5a) using the spectrum The first spectral component in DC component right Perform an inverse fast Fourier transform to calculate the DC component λ of the profile. D : (5b) Spectrum The first spectral component in Setting the value to 0 yields a spectrum consisting of q spectral components that do not contain a DC component. For the spectrum After performing an N-point inverse fast Fourier transform and taking its real part, a contour λ of length N without DC component is obtained. N : in and Spectrum and spectrum The (m+1)th spectral component, where N is a positive integer, and Re is the operation for finding the real part of the complex number; (5c) Through the profile λ that does not contain a DC component N DC component λ of the profile D Calculate the profile λ of the X-ray pulsar radiation signal: λ=λ N +λ D 。
Citation Information
Patent Citations
X-ray pulsar navigation ground verification system based on visible light source
CN108680187A
X-ray pulsar short-time observation high-signal-to-noise ratio contour acquisition method
CN110986922A
