A method for extracting a mixed observation signal of multiple pulsars
By using a periodic coincidence algorithm and a higher-order correlation statistical model, the target pulsar signal is separated and extracted from the mixed signal of multiple pulsars, solving the problem that traditional methods cannot remove noisy photons, and realizing efficient pulsar signal processing and navigation positioning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG INST OF AEROSPACE ELECTRONICS TECH
- Filing Date
- 2023-07-31
- Publication Date
- 2026-07-21
AI Technical Summary
Existing technologies struggle to effectively remove noise photons and extract the target pulsar signal from mixed signals from multiple pulsars. Traditional filtering methods cannot eliminate noise photons within the same line-of-sight and energy spectrum range, nor can they separate signals from multiple pulsars.
The algorithm employs a pulsar period search module, a target pulsar information separation module, a pulsar signal and noise separation module, and a noise suppression evaluation module. Through periodic coincidence algorithm and high-order correlation statistical model, it separates and extracts pulsar signal photons, removes noise photons, and utilizes the periodicity of pulsar signals and the differences between noise photons to achieve signal and noise separation.
It achieves effective extraction of target pulsar signals and effective removal of noise photons from mixed signals of multiple pulsars, improving the accuracy and reliability of navigation and positioning, and achieving a noise removal effect of 91.94%-92.1%.
Smart Images
Figure CN117091609B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of X-ray pulsar navigation and signal processing technology, specifically relating to a method for extracting mixed observation signals from multiple pulsars. Background Technology
[0002] Pulsar navigation is a novel method of autonomous navigation for spacecraft. It calculates the spacecraft's position relative to the solar system's center of mass by measuring the time difference between the arrival time (tsc) of pulsar photons and their arrival time (tssb) at the solar system's center of mass. By measuring multiple pulsars (three or more), comprehensive navigation and positioning of the spacecraft can be achieved.
[0003] Limited by spacecraft resource requirements, simultaneous observation of multiple pulsar signals over a wide field of view is an effective way to achieve pulsar navigation applications, based on the needs of real-time navigation. However, this approach presents significant challenges in noise removal and target pulsar extraction. Traditional pulsar signal processing algorithms often employ filtering methods, but these methods can only remove cosmic background noise and cannot eliminate noise photons in the same line-of-sight and energy spectrum range as the pulsar signal. Furthermore, they cannot extract the target pulsar signal from a mixture of multiple pulsar signals. Summary of the Invention
[0004] To address the problems existing in the background art, the present invention provides a method for extracting mixed observation signals from multiple pulsars, which includes:
[0005] A pulsar period search module for determining the period information of each pulsar in a mixed signal from multiple pulsars;
[0006] A target pulsar information separation module for separating and extracting information from different pulsars in mixed information;
[0007] A pulsar signal and noise separation module that enables the separation and extraction of signal photons and noise photons from a target pulsar;
[0008] The isolated pulsar signal photons are outlined and compared with a standard outline to obtain a standard outline comparison module for outline similarity.
[0009] A noise suppression evaluation module analyzes the target pulsar signal after periodic coincidence processing to evaluate the noise suppression effect of this technique.
[0010] The following steps are required:
[0011] S1. The mixed signal from multiple pulsars first passes through the pulsar period search module. Based on the prior period information of the target pulsar, the optimal period information of the current target pulsar is determined by frequency search to eliminate the period drift of the target pulsar signal caused by the period instability of the pulsar and the interference of the test process.
[0012] S2. Following the method in S1, determine the period information of the remaining pulsars in the mixed signal;
[0013] S3. After determining the periodic information of the target pulsar, the mixed signal of multiple pulsars enters the pulsar information separation module. For the periodic characteristics of different pulsars, the corresponding coincidence width and periodic coincidence order are set to achieve the separation of information from different pulsars.
[0014] S4. After information separation, a pulsar signal and noise separation module is used to separate the photons and noise of the target pulsar signal.
[0015] Furthermore, the specific process of step S3 includes:
[0016] Q1. First, extract the pulsars with strong flux to avoid interference caused by extracting information from other pulsars;
[0017] Q2. For high-flow pulsar signals with a period of Tp, the coincidence width is selected near the period based on the periodic characteristics of the signal. The initial value of the coincidence width is selected as Tp+Δt, and the initial coincidence threshold is selected based on the number of photons of the detected target pulsar.
[0018] Q3. Using a periodic coincidence algorithm, with an initial coincidence width, the coincidence value is incremented by 1 for each photon from the target pulsar that appears, until the coincidence threshold is reached.
[0019] Q4. After the initial extraction, a frequency detection algorithm is used to detect whether the frequency components of the target pulsar signal still exist in the mixed signal. If so, step Q3 is repeated, and the coincidence width and coincidence threshold are updated until the frequency components of the target pulsar signal no longer appear in the mixed signal, thus completing the extraction of the pulsar signal.
[0020] Q5. Repeat steps Q1-Q4 to complete the extraction of the remaining pulsar information.
[0021] Furthermore, the specific process of step S4 includes:
[0022] H1. The signal from the target pulsar is divided into two equal segments. The initial coincidence width is determined based on the difference in the higher-order correlation statistical model. The initial coincidence threshold is set to the number of signal photons.
[0023] H2. Determine whether a photon exists simultaneously at the beginning of both signal segments within the initial coincidence width. If it exists, the photon is considered a signal photon, and the coincidence count is incremented by 1. If they do not exist simultaneously, the signal photon is considered not to exist at that position.
[0024] H3. After the initial judgment, the window is sequentially cyclically slid according to the signal period, and the judgment is performed according to the method in H2 until the count meets the threshold requirement.
[0025] H4, Components: If steps H2-H3 are repeated, and the coincidence width and coincidence threshold are updated, until the frequency components of the target pulsar signal no longer appear in the mixed signal, the extraction of the pulsar photon signal is completed.
[0026] The beneficial effects achieved by this invention are as follows:
[0027] First, this invention utilizes the unique periodicity of pulsar signals to propose a signal separation method for different pulsars based on a combination of periodic coincidence and high-precision frequency estimation. By changing the coincidence order and using high-precision frequency estimation, the method effectively extracts target pulsar information from mixed pulsar information of multiple pulsars, providing technical support for simultaneous observation of multiple pulsars by a single detector and subsequent pulsar navigation applications.
[0028] Secondly, this invention utilizes the characteristic differences between noise photons and signal photons to propose a pulsar noise removal module and method based on coincidence. By periodically sliding the coincidence window, it achieves effective separation of pulsar signal photons and noise photons, solving the problem that traditional pulsar denoising methods cannot remove noise photons that are in the same line of sight and energy spectrum range as the pulsar. Attached Figure Description
[0029] Figure 1 The diagram shows a technical scheme for extracting multi-source mixed signals from pulsars.
[0030] Figure 2 A schematic diagram illustrating the 5x noise level of Crab in implementation 1;
[0031] Figure 3 A schematic diagram illustrating the 5x noise level for implementing B1509 in section 1;
[0032] Figure 4 A schematic diagram illustrating the 5x noise level of J1814 in implementation 1;
[0033] Figure 5 This is the Crab signal extraction diagram in implementation 1;
[0034] Figure 6 Crab analysis plot of the remaining mixed signal in Implementation 1;
[0035] Figure 7 This is the signal extraction diagram of B1509 in Implementation 1;
[0036] Figure 8 This is the analysis diagram of the remaining mixed signal B1509 in Implementation 1;
[0037] Figure 9 This is the signal extraction diagram of J1814 in Implementation 1;
[0038] Figure 10 This is the analysis diagram of the remaining mixed signal J1814 in Implementation 1. Detailed Implementation
[0039] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. In addition, the forms of the various structures described in the following embodiments are merely illustrative. The present invention is not limited to the structures described in the following embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0040] This invention provides a method for extracting mixed observation signals from multiple pulsars. This method comprises five modules: a pulsar period search module, a target pulsar information separation module, a pulsar signal and noise separation module, a standard profile comparison module, and a noise suppression evaluation module.
[0041] The main function of the pulsar period search module is to determine the period information of each pulsar in a mixed signal of multiple pulsars. The period can be determined using a chi-square search. Given an initial period value and search range for a single pulsar, the period corresponding to the maximum chi-square value is the optimal period for that pulsar.
[0042] The main function of the pulsar information separation module is to separate and extract information from different pulsars in mixed signals. The key innovation of this module lies in utilizing the unique periodic characteristics of different pulsar signals to propose a periodic coincidence algorithm suitable for separating mixed pulsar signals. By setting different coincidence widths and periodic coincidence orders for different pulsars, the information of the target pulsar can be extracted. The extracted target pulsar information includes both signal photons and noise photons.
[0043] The main function of the pulsar signal and noise separation module is to separate and extract the target pulsar signal photons and noise photons. The main innovation of this module is to utilize the difference in the high-order correlation statistical model between pulsar signal photons and noise photons to further optimize the coincidence window and coincidence order, thereby achieving the removal of 90% of noise photons.
[0044] The standard contour comparison module folds the isolated pulsar signal photons into contours and compares them with standard contours to obtain contour similarity.
[0045] The main function of the noise suppression evaluation module is to analyze the target pulsar signal after periodic coincidence processing and evaluate the noise suppression effect of the technique.
[0046] The specific implementation process is as follows:
[0047] First, the mixed signal from multiple pulsars passes through a pulsar period search module. Based on the prior period information of the target pulsar, the optimal period information of the current target pulsar is determined through frequency search to eliminate period drift of the target pulsar signal caused by pulsar period instability and interference during the testing process. The period information of the remaining pulsars in the mixed signal is determined using the same method.
[0048] 2. After determining the periodicity of the target pulsar, the mixed signal from multiple pulsars then enters the pulsar information separation module. For different pulsar periodic characteristics, corresponding coincidence widths and periodic coincidence orders are set to separate the information from different pulsars. The specific workflow in this module is as follows:
[0049] (1) When designing the extraction algorithm, the pulsars with strong flow are extracted first, because the strong flow pulsars account for a large proportion in the mixed signal. Extracting them first can avoid interference caused by extracting the information of other pulsars.
[0050] (2) For high-flow pulsar signals with a period of Tp, when designing the extraction algorithm, the coincidence width should be selected near the period according to the periodic characteristics of the signal. The initial value of the coincidence width is selected as Tp+Δt, and the initial coincidence threshold is selected based on the number of photons of the detected target pulsar (including signal and noise).
[0051] (3) A periodic coincidence algorithm is adopted. Under the initial coincidence width, the coincidence value is incremented by 1 for each photon of the target pulsar until the coincidence threshold requirement is reached.
[0052] (4) After the first extraction, a frequency detection algorithm is used to detect whether the frequency component of the target pulsar signal still exists in the mixed signal. If so, step (3) is repeated and the conformity width and conformity threshold are updated until the frequency component of the target pulsar signal no longer appears in the mixed signal, thus completing the extraction of the pulsar signal.
[0053] (5) Repeat the above steps to complete the extraction of the remaining pulsar information.
[0054] II. After information separation, a pulsar signal and noise separation module is used to separate the target pulsar signal photons from the noise. This process mainly utilizes the difference in the higher-order correlation statistical models of pulsar signal photons and noise photons to design a suitable periodic coincidence algorithm for the pulsar signal photons to achieve the separation of signal photons and noise. For pulsar signal photons, due to their periodic characteristics, their higher-order correlation statistical model is a periodic thermal-optical field distribution, while the higher-order correlation statistical model of noise photons is an aperiodic thermal-optical field distribution. The difference between the two can be used as the basis for pulsar signal extraction. Based on the above analysis, the specific analysis process of this module is as follows.
[0055] (1) The signal of the target pulsar is divided into two segments. The initial coincidence width is determined based on the difference of the higher-order correlation statistical model. The initial coincidence threshold is set to the number of signal photons.
[0056] (2) Determine whether a photon exists simultaneously at the beginning of the two signal segments within the initial coincidence width. If it exists, the photon is considered a signal photon, and the coincidence count is incremented by 1. If they do not exist simultaneously, the signal photon is considered not to exist at that position.
[0057] (3) After the initial judgment is completed, the window is sequentially cyclically slid according to the period of the signal, and the judgment is performed according to the method in (2) until the count meets the threshold requirement.
[0058] (4) After the first extraction, the frequency detection algorithm is used to detect whether there are still frequency components of the pulsar photon signal in the signal. If there are, repeat steps (2)-(3) and update the conformity width and conformity threshold until the frequency components of the target pulsar signal no longer appear in the mixed signal, thus completing the extraction of the pulsar photon signal.
[0059] Fourth, the isolated pulsar signal photons are contoured to obtain test contours, and the similarity of the contours is analyzed using a standard contour comparison module. Similarity calculation can be performed by cross-correlation analysis of the two contours; the maximum cross-correlation value is the similarity between the two contours.
[0060] Example 1,
[0061] The effects of the present invention will be explained below with reference to specific implementation examples.
[0062] To verify the performance of the method proposed in this patent, an analysis example of signal extraction from the mixed information of three pulsars using the method proposed in this patent is presented. The conditions for this example are shown in Table 1:
[0063] Table 1 Simulation parameters of the mixed signal from the three pulsars
[0064] Simulation parameters Crab B1509-58 J1814-338 Period (ms) 33.729 150.23 3.18 Number of signal photons (ph) 15400 1620 3880 Noise level (ph) 154000 16200 38800
[0065] Figures 2-4 The original signals from pulsars Crab, B1509, and J1814, along with five times the noise, were mixed. Then, a multi-source pulsar signal extraction technique was used to extract the signal contours of each pulsar, as shown below. Figure 5 , Figure 7 and Figure 9 As shown, this method can effectively extract signal photons from target pulsars.
[0066] Simultaneously, the extracted mixed signal is analyzed, such as... Figure 6 , Figure 8 and Figure 10 As shown, the extracted target pulsar signal can no longer be recovered from the mixed signal, indicating that the method can extract effective photon signals and remove noisy photon signals.
[0067] In this embodiment, by setting different flag bits for the signal and noise of the target pulsar, and analyzing the flag bits, it can be concluded that this method can effectively extract the information of the target pulsar. The noise removal effects for the three pulsars Crab, B1509, and J1814 are 91.94%, 92.1%, and 91.5%, respectively.
[0068] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for extracting mixed observation signals from multiple pulsars, characterized in that, Its settings include: A pulsar period search module for determining the period information of each pulsar in a mixed signal from multiple pulsars; A target pulsar information separation module for separating and extracting information from different pulsars in mixed information; A pulsar signal and noise separation module that enables the separation and extraction of signal photons and noise photons from a target pulsar; The isolated pulsar signal photons are outlined and compared with a standard outline to obtain a standard outline comparison module for outline similarity. A noise suppression evaluation module analyzes the target pulsar signal after periodic coincidence processing to evaluate the noise suppression effect of this technique. The following steps are required: S1. The mixed signal from multiple pulsars first passes through the pulsar period search module. Based on the prior period information of the target pulsar, the optimal period information of the current target pulsar is determined by frequency search to eliminate the period drift of the target pulsar signal caused by the period instability of the pulsar and the interference of the test process. S2. Following the method in S1, determine the period information of the remaining pulsars in the mixed signal; S3. After determining the periodic information of the target pulsar, the mixed signal of multiple pulsars enters the pulsar information separation module. For the periodic characteristics of different pulsars, the corresponding coincidence width and periodic coincidence order are set to achieve the separation of information from different pulsars. S4. After information separation, a pulsar signal and noise separation module is used to separate the photons and noise of the target pulsar signal.
2. The method for extracting mixed observation signals from multiple pulsars according to claim 1, characterized in that: The specific process of step S3 includes: Q1. First, extract the pulsars with strong flux to avoid interference caused by extracting information from other pulsars; Q2. For high-flow pulsar signals with a period of Tp, the coincidence width is selected near the period based on the periodic characteristics of the signal. The initial value of the coincidence width is selected as Tp+Δt, and the initial coincidence threshold is selected based on the number of photons of the detected target pulsar. Q3. Using a periodic coincidence algorithm, with an initial coincidence width, the coincidence value is incremented by 1 for each photon from the target pulsar that appears, until the coincidence threshold is reached. Q4. After the initial extraction, a frequency detection algorithm is used to detect whether the frequency components of the target pulsar signal still exist in the mixed signal. If so, step Q3 is repeated, and the coincidence width and coincidence threshold are updated until the frequency components of the target pulsar signal no longer appear in the mixed signal, thus completing the extraction of the pulsar signal. Q5. Repeat steps Q1-Q4 to complete the extraction of the remaining pulsar information.
3. The method for extracting mixed observation signals from multiple pulsars according to claim 1, characterized in that: The specific process of step S4 includes: H1. The signal of the target pulsar is divided into two segments. The initial coincidence width is determined based on the difference in the higher-order correlation statistical model. The initial coincidence threshold is set to the number of signal photons. H2. Determine whether photons exist simultaneously at the beginning of the two signal segments within the initial coincidence width. If they do, the photon is considered a signal photon, and the coincidence count is incremented by 1. If they do not exist simultaneously, the signal photon is considered not to exist at that position. H3. After the initial judgment, the window is cyclically slid according to the signal period, and the judgment is performed according to the method in H2 until the count meets the threshold requirement. H4, Components: If steps H2-H3 are repeated, and the coincidence width and coincidence threshold are updated, until the frequency components of the target pulsar signal no longer appear in the mixed signal, the extraction of the pulsar photon signal is completed.