Space decoupling method for preliminary search of geosynchronous orbit system
Patent Information
- Application Number
- CN202610213850.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2025-02-28
- Filing Date
- 2026-02-13
- Publication Date
- 2026-08-28
AI Technical Summary
[0006]The main effects and advantages of this invention are that it can reduce search complexity, improve search efficiency, enhance peak position detection accuracy, avoid power dead zone problems, and is applicable to dynamic communication scenarios.
Smart Images
Figure CN122660700A_ABST
Abstract
Description
[0001] This application claims priority to U.S. Patent Application No. 63 / 765,434, filed on February 28, 2025, the entire contents of which are incorporated herein by reference. Technical Field
[0002] This invention relates to the field of satellite communications, and in particular to a space decoupling method for the initial search of a geosynchronous Earth Orbit (GEO) system. Background Technology
[0003] Satellite communications (SATCOM) transmits signals between a user terminal located on the Earth's surface and a target satellite in Earth orbit via electromagnetic beams. The user terminal includes both fixed and mobile communication equipment, including but not limited to equipment mounted on moving platforms such as ships and vehicles on the Earth's surface. Beam alignment between the user terminal's antenna and the target satellite is crucial for ensuring accurate link transmission. To ensure the accuracy of link transmission, these user terminals must align their beams with the target satellite.
[0004] The user terminal's antenna uses beam directivity to focus the signal onto the target satellite, achieving precise connection and transmission. When the user terminal is in motion, it needs to search for the target satellite in geostationary orbit and adjust its beam alignment accordingly. Summary of the Invention
[0005] The main objective of this invention is to provide a space decoupling method for the initial search of geosynchronous orbit systems. To achieve the aforementioned objective, this invention adopts the following technical solution: A space decoupling method for preliminary searches of geosynchronous orbit systems includes the following steps performed sequentially: Step 1: Analyze and receive the spatial selectivity of the beamforming output, and decouple at least one one-dimensional decoupled search space from the joint / complete three-dimensional search space; Step 2: Based on the peak position guess from the decoupled search space, establish and form the corresponding two-dimensional reduced search space, and perform the first peak position detection in the reduced search space; Step 3: Based on the detection results of the first peak position detection, perform second peak position detection in the decoupled search space; and Step 4: Integrate the peak position detection results from the decoupled search space and the dimensionality-reduced search space to reconstruct the final peak position.
[0006] The main effects and advantages of this invention are that it can reduce search complexity, improve search efficiency, enhance peak position detection accuracy, avoid power dead zone problems, and is applicable to dynamic communication scenarios. Attached Figure Description
[0007] Figure 1 This displays an explanation of the spatial components pointing to the mechanical antenna in the three-dimensional search space.
[0008] Figure 2 Displays the definition of the spatial components pointing to the antenna in three-dimensional space.
[0009] Figure 3 Displays the search candidate set in a given search space.
[0010] Figure 4 This demonstrates the spatial decoupling mechanism of the present invention.
[0011] Figure 5 This invention demonstrates the detection mechanism for multi-stage peak position detection.
[0012] Figure 6 This displays the beamforming output power distribution along the skew axis.
[0013] Figure 7 The waveform of the received beamforming output power is displayed, showing the angle jumps.
[0014] Figure 8 Display space decoupling enables dimensionality reduction of the search space.
[0015] Figure 9 This diagram shows the peak position detection of the present invention.
[0016] Figure 10 This is the initial waveform shaping output power map, showing the beamforming output power distribution along the azimuth axis.
[0017] Figure 11 This is the initial waveform shaping output power map, showing the beamforming output power distribution along the downtilt axis.
[0018] Figure 12 This is the initial waveform shaping output power diagram, showing the beamforming output power distribution along the skew axis.
[0019] Figure 13 A diagram showing the decoupling of the display space into two search spaces.
[0020] Figure 14 This demonstrates an example of spatial decoupling into two search spaces when spatial variation along a certain direction is almost constant.
[0021] Figure 15This demonstrates an example of spatial decoupling into two search spaces when spatial variation along a certain direction is almost constant.
[0022] Figure 16 The reduced-dimensional search space is shown and separated from the three-dimensional full search space.
[0023] Figure 17 Display the decoupled search space, which is separate from the three-dimensional full search space.
[0024] Figure 18 This illustrates another embodiment of the spatial decoupling of the present invention.
[0025] Figure 19 This illustrates another embodiment of the spatial decoupling of the present invention.
[0026] Figure 20 This is the initial beamforming output power diagram from the search, showing... The uniform linear array generates output power in a two-dimensional beamforming pattern between the azimuth and downtilt axes.
[0027] Figure 21 This is the initial beamforming output power diagram from the search, showing... The uniform linear array generates output power in a two-dimensional beamforming pattern between the azimuth axis and the downtilt axis.
[0028] Figure 22 This is the initial beamforming output power diagram from the search, showing... The uniform linear array generates output power in a two-dimensional beamforming pattern between the azimuth axis and the downtilt axis.
[0029] Figure 23 This is a preliminary beamforming output power map, showing an example of peak location detection based on two-dimensional beamforming output power between the azimuth and downtilt axes.
[0030] Figure 24 This is a preliminary beamforming output power map, showing an example of peak location detection based on two-dimensional beamforming output power between the azimuth and skew axes (peaks are not identifiable).
[0031] Figure 25 This is a preliminary beamforming output power map, showing an example of peak location detection based on two-dimensional beamforming output power between the downtilt axis and the skew axis (peaks cannot be identified).
[0032] Figure 26 This is a preliminary beamforming output power map, showing an example of peak position detection based on two-dimensional beamforming output power between the azimuth and downtilt axes (peak position remains unchanged).
[0033] Figure 27This is a preliminary beamforming output power map, showing an example of peak position detection (peak position drift) based on two-dimensional beamforming output power between the azimuth and skew axes.
[0034] Figure 28 This is a preliminary beamforming output power map, showing an example of peak position detection (peak position drift) based on two-dimensional beamforming output power between the downtilt axis and the skew axis.
[0035] Figure 29 This is a preliminary beamforming output power map, showing an example of peak position detection based on two-dimensional beamforming output power between the azimuth and downtilt axes (peak position remains unchanged, power decreases slightly).
[0036] Figure 30 This is a preliminary beamforming output power map, showing an example of peak position detection (peak position drift) based on two-dimensional beamforming output power between the azimuth and skew axes.
[0037] Figure 31 This is a preliminary beamforming output power map, showing an example of peak position detection (peak position drift) based on two-dimensional beamforming output power between the downtilt axis and the skew axis.
[0038] Figure 32 This is a preliminary waveform shaping output power diagram, showing an example of beamforming output power distribution along the azimuth axis.
[0039] Figure 33 This is a preliminary waveform shaping output power diagram, showing an example of beamforming output power distribution along the downtilt axis.
[0040] Figure 34 This is a preliminary waveform shaping output power diagram, showing an example of beamforming output power distribution along the skew axis.
[0041] Figure 35 This is the initial waveform shaping output power diagram, showing the random guessing process of the power dead zone problem.
[0042] Figure 36 Displays a description of the array index mapping. Detailed Implementation
[0043] The diagram illustrates an embodiment of the space decoupling method for the initial search of the geosynchronous orbit system according to the present invention. It is for illustrative purposes only and is not intended to limit the scope of the invention.
[0044] The optimal search for the three-dimensional joint space using a full-angle / full-domain search method is called 3D main beam acquisition. 3D main beam acquisition is a brute-force (exhaustive) blind search (scanning) mechanism. This method can probe one axis at a time or alternate between the three axes. It is a simple, low-cost, and time-consuming method whose performance depends on either the step size or the signal-to-noise ratio (SNR), and requires no prior information. The unified exhaustive search two-stage strategy involves: in the first stage, large-scale probing is performed using a fixed, relatively large step size; then, in the second stage, small-scale probing is performed around the search point from the first stage using a fixed, relatively small step size. When the pointing error is within 3 dB of the beamwidth, the search stopping criterion is met, and the search stops, thus achieving the minimum link level required for alignment (control signal). For example: ,in, To receive the output power of the beam, To receive the minimum required output power, The output power is the output power at a 3 dB receive beamwidth.
[0045] like Figure 1 As shown, the problem with the direct exhaustive blind search (scanning) mechanism (linear / uniform search) lies in the exhaustive peak location detection (PLD) of the received (Rx) beamforming output. The objective function is the received beamforming output power and is also an antenna pointing function, which is a... A ternary function, where This indicates that the antenna of a user terminal located on the Earth's surface is oriented around the direction of movement. The tilt angle of rotation, This indicates that the antenna surrounds the side. The downward tilt angle of rotation, This indicates that the antenna is oriented in the direction of gravity. The azimuth angle of rotation. The goal is to find the output power that enables receiver beamforming. Maximize antenna pointing Based on a three-dimensional orthogonal standard basis. The peak location (PL) in the joint / complete 3D search space can be determined through the corresponding The peak position is estimated based on the joint objective function. The measurements were obtained in the combined / complete three-dimensional space. .
[0046] like Figure 2 As shown, in the joint / complete 3D search space, uniform peak position detection is performed on all possible search candidate points. Therefore, the set of all candidate points is called the uniform search candidate set. Here, "uniform" means that the distribution of search candidate points is uniform, which is defined by the specified starting position and step size. The mathematical expression of the joint / complete 3D search space is:
[0047] in ; Represents the joint / complete three-dimensional search space (three-dimensional, ); Represents the first in the joint / complete three-dimensional search space The orthogonal basis (specified as the nth) Search directions ); Indicates the first The number of the standard basis (specifying the first) is the number of the standard basis. (Search locations).
[0048] like Figure 3 As shown, the uniform (linear) search is defined as follows:
[0049] in Represents a 3D uniform search candidate set in the joint / complete 3D search space, containing a total of Search candidate points of 100 possible candidates ,and Each search candidate point Represented as:
[0050] in ; For the Search directions :
[0051] in For along the first Search directions A uniform search set containing candidate sets , ; Search parameters: Search range: along the first Search range Its range is defined by the Field of View (FoV). Decide:
[0052] Step size: along the first Search directions The constant step size (i.e., the search interval) is expressed as: ; Number of candidates: Search directions The number of candidates (i.e., the size of the search set) Expressed as: .
[0053] Specifically, Figure 3 Displayed in the third search direction Search candidate set The search range is =180°, step size is Δ =1°, therefore the number of candidate points is =180, search range from -90° to +90°.
[0054] Total number of candidate points in the joint / complete 3D search space The total number of searches is given by the following formula:
[0055] For example: ;
[0056] That is, the total number of candidate points is the direct product of the number of candidate points in each search direction.
[0057] Total number of searches in the joint / complete 3D search space (i.e., total number of candidate points) The calculation is as follows:
[0058] Example 1: , , , : ; Example 2: , , , (For a 32-element array, the 3 dB beamwidth at azimuth / downtilt is approximately) ): .
[0059] Although the direct exhaustive blind search (scanning) mechanism is obvious and can achieve optimal results in mechanical antenna orientation mechanisms, the large number of searches required leads to significant computational time and is accompanied by wear and tear on the mechanical structure.
[0060] A suboptimal solution suggests using spatial decoupling techniques to reduce the original search space to a lower-dimensional search space. Generally, this can be achieved by establishing... An independent search space is proposed to reduce the number of searches, and a multi-stage peak position detection (MSPLD) method is also proposed, which uses a progressively decreasing search step size. Generally speaking, This can be achieved in one stage, further reducing computational complexity and accelerating decision-making time. Additionally, the angle jump process (generally speaking) can be simplified. This single jump process can avoid power dead zones during peak position detection, thereby improving overall system performance; the aforementioned This indicates the number of lower-dimensional search spaces formed by decoupling. This indicates the number of stages in the multi-stage peak position detection. This represents the number of jumps in an angular jump.
[0061] like Figures 4 to 7 As shown, the present invention proposes a suboptimal solution for the initial search (cold start phase) of beam alignment of an antenna to a target satellite in geosynchronous orbit, including the following key operations: (1) spatial decoupling to reduce the search space dimension; (2) multi-stage peak position detection to speed up processing time; and (3) angle skipping in peak position detection to avoid power dead zone problem.
[0062] like Figure 8 and Figure 9 As shown, the spatial decoupling process used for search space dimensionality reduction includes the following steps performed sequentially: Step 1 (Decoupling): Analyze the spatial selectivity of the received (Rx) beamforming output and perform decoupling; Step 2 (Detection of the first peak position in the dimensionality reduction search space): Detection of the first peak position in the dimensionality reduction search space is performed by using a given peak position guess from the decoupled search space; Step 3 (Second Peak Position Detection in the Decoupled Search Space): Using the first peak position detection obtained from the dimensionality-reduced search space, perform second peak position detection in the decoupled search space; and Step 4 (Reconstruction): Reconstruct the peak detection based on the results of the second peak position detection from the decoupled search space and the first peak position detection from the dimensionality-reduced search space.
[0063] The following provides a more detailed explanation of the four spatial decoupling processes mentioned above.
[0064] Step 1: As Figure 10 , Figure 11 and Figure 12 As shown, step one analyzes the spatial selectivity of beamforming output power and then performs spatial decoupling. The spatial selectivity analysis is based on the beam pointing pattern on an analog (mechanical) angle scale, rather than on a discrete (electrical) angle scale. The spatial selectivity (i.e., angle variation characteristics) is determined by the spatial degrees of freedom (DoF), which in turn are determined by the antenna array size degree of freedom and polarization degree of freedom. Figure 10 Display targeting Spatial degrees of freedom It is determined by the size of the two-dimensional antenna array Created Figure 11 Display targeting Spatial degrees of freedom It is determined by the size of the two-dimensional antenna array Created Figure 12 Display targeting Spatial degrees of freedom It is created by the antenna polarization degrees of freedom; in Figure 12 In China, due to The degree of freedom of direction is lower than and The degree of freedom of direction, therefore along The direction observed is a very slow "angle change".
[0065] Spatial decoupling: If the spatial selectivity of the beam output power is along the first... direction If the degree of change is relatively very small, then it has at least one direction. From the base The extended search space Can be found in a joint / complete three-dimensional search space It is decoupled and used for peak position detection based on beam output power.
[0066] Corollary 1: The joint / complete 3D search space can be decoupled into more than one (i.e., A separate (i.e. nearly independent) search space that can be used for peak location detection.
[0067] Corollary 2: When along the first Search directions When the spatial variation is almost constant, the corresponding search space This allows for separation from the joint / complete 3D search space, i.e.: It is approximately equal to a constant.
[0068] Inference 3: Random guessing The estimated value can be obtained through Arbitrarily pre-specified (thus decoupling) without significantly altering other directions. Beamforming output power The beamforming output power in these directions is defined as:
[0069] in Indicated as by ( The received beamforming output power within the extended space. For example, in two-dimensional space: .
[0070] Inference 4: The main beam profile will be along the first Search directions It exhibits a near-linear distribution, resembling a simple straight line.
[0071] Inference 5: Generally speaking, This represents two independent search spaces: a dimensionality-reduced search space and a decoupled search space.
[0072] like Figure 13 As shown, Definition 1: Any one-dimensional search space that can be decoupled from the joint / complete three-dimensional search space (with relatively low spatial selectivity) is defined as a “decoupled” space, and the remaining two-dimensional joint space after decoupling from the initial joint / complete three-dimensional search space is defined as a “reduced” space.
[0073] like Figure 14 As shown, spatial decoupling: at least one along the first Search directions search space (by basis vectors) The extended (then) can be approximately decoupled from the joint / complete three-dimensional search space if the beamforming output power is along the search direction. If the spatial variation is small enough, then the decoupled search space (denoted as ) It can be used for peak position detection based on beamforming output power.
[0074] Step 2: Perform peak location detection in the dimensionality-reduced search space. Based on the spatial decoupling described above, such as... Figure 15 As shown, it can be derived from the joint objective function Obtain equivalent dimensionality-reduced receiver beamforming output power Due to the joint objective function variables in It can be arbitrarily specified in advance, without affecting the joint objective function. It has a significant impact, so they can be almost derived from the joint objective function. Decoupling and separation in the middle. Among them, it has... Variables Dimensional reduction receiver beamforming output power It is a joint objective function The subfunction, therefore .
[0075] like Figure 16 As shown, if along the first Search directions (by basis vectors) The search space (defined) If it can be discarded or separated from the joint / complete three-dimensional search space, a two-dimensional uniform search set can be constructed. A two-dimensional uniform search set is composed of a three-dimensional uniform search set. This is derived from dimensionality reduction. Specifically, the first [dimensionality] in the two-dimensional uniform search set... The candidate is represented as follows:
[0076] Here, the mapping between indices is given by the following formula: ,
[0077] In the dimensionality reduction search space In the two-dimensional uniform search set Number of candidates It is given by the following formula:
[0078] in, Indicates the first A dimensionality reduction search space Middle of the first Number of candidate sets for each search direction.
[0079] The estimation of the peak position is done in a dimensionality-reduced search space. Two-dimensional uniform search set It will be carried out internally.
[0080] Step 3: As Figure 17 As shown, peak position detection is performed in the decoupled search space, which has an almost flat angular profile. Peak position detection is performed in a lowest priority order, where in step two, arbitrarily assigned temporary estimates are distributed along the decoupled search direction to facilitate dimensionality reduction through spatial decoupling. While randomly selected temporary estimates can speed up the decoupling process, they may not provide optimal performance. Therefore, after spatial decoupling, accurate peak positions must be re-estimated in step three to improve performance. The decoupled one-dimensional decoupled search space is defined as follows: In its first From each direction constitute:
[0081] In the decoupled search space In the middle, a one-dimensional uniform search set Candidate number Determined by the following formula: In the direction of the off-couple search Above, re-estimate the uniform search candidate set. The peak position within.
[0082] Step Four: As Figure 18 As shown, spatial decoupling techniques are used to reduce the dimensionality of the search space. and the search space for couples By integrating the peak estimates from the data, joint / complete 3D peak locations can be reconstructed. .
[0083] The total number of peak location detection searches is calculated as follows:
[0084] If peak position detection is used to combine / complete the three-dimensional search space Decomposed into and Composition of two-dimensional joint search space and by One-dimensional search space The total number of searches across all individual search spaces can then be determined as follows: 1. For the first individual search space : .
[0085] 2. For the second separate search space : .
[0086] Example 3: For a joint / complete 3D search space Using the conventional method, if the search parameters are: , Then we can get .
[0087] Example 4: Consider detection based on spatial decoupling and dimensionality reduction, combining / complete three-dimensional search space. Decomposed into orthogonal standard basis and Composition of two-dimensional joint search space and by orthogonal standard basis One-dimensional search space If the search parameters are as follows: , , , Then we can get ; .
[0088] Next, we will discuss some practical issues related to spatial decoupling in the initial search. For example... Figure 19 As shown, the first question is whether the joint / complete 3D search space can be completely decoupled into... Is there a maximum number of independent one-dimensional search spaces? The answer is no, because typically, the maximum number of separable search spaces is 1 / 2. .
[0089] If the joint / complete 3D search space is completely decoupled into A one-dimensional (1D) independent and separate decoupled search space (direction). (in The detection process is as follows: A candidate set is searched uniformly in one dimension. The peak position is detected independently in the direction of the rear. In the middle, for other search directions ( Use a given temporary random guess or a pre-specified estimate. Finally, the three independent one-dimensional peak position detections are combined to reconstruct the final joint three-dimensional peak position.
[0090] However, as Figure 20 , Figure 21 and Figure 22 As shown, complete decoupling is impossible for geosynchronous orbit systems. In fact, the joint / complete three-dimensional search space can only be decoupled into two independent search spaces: one based on the orthogonal standard basis. and orthogonal standard basis Composition of two-dimensional joint search space and by orthogonal standard basis One-dimensional search space However, based on orthogonal standard basis... and orthogonal standard basis Composition of two-dimensional joint search space It is impossible to further completely decouple them into two independent entities, each based on an orthogonal standard basis. and orthogonal standard basis The search space formed remains the same even if the array size along one axis is significantly smaller than the array size along the other axis. For example... Figure 21 and Figure 22 As shown, the main beam profile does not exhibit a simple linear distribution along any particular search datum.
[0091] The following example illustrates the practical problem of spatial decoupling between the dimensionality reduction search space and the decoupled search space.
[0092] Example 5: such as Figure 23 As shown, the joint / complete 3D search space is decoupled into two independent spaces, in which... and Using true estimates in the extended space To estimate The peak value is obvious.
[0093] Example 6: such as Figure 24 As shown, the joint / complete 3D search space is separated into two independent spaces, in which... and The extended space uses true estimates To estimate The peak detection probability decreases as the signal-to-noise ratio decreases, therefore peaks cannot be distinguished.
[0094] Example 7: such as Figure 25 As shown, the joint / complete 3D search space is separated into two independent spaces, in which... and The extended space uses true estimates estimate The peak detection probability decreases as the signal-to-noise ratio decreases, therefore peaks cannot be distinguished.
[0095] Example 8: such as Figure 26 As shown, the joint / complete 3D search space is separated into two independent spaces, in which... and Use in the extended space To estimate by random guessing Compared to Example 5, the beam pointing pattern is slightly distorted, but the peak position remains unchanged.
[0096] Example 9: For example Figure 27 As shown, the joint / complete 3D search space is separated into two independent spaces, in which... and Use in the extended space To estimate by random guessing Compared to Example 6, the beam pointing pattern is significantly distorted, resulting in a shift in the peak position.
[0097] Example 10: (The rest of the text appears to be a fragment and doesn't form a coherent sentence or paragraph.) Figure 28 As shown, the joint / complete 3D search space is separated into two independent spaces, in which... and Used in the extended space To estimate by random guessing Compared to Example 7, the beam pointing pattern is significantly distorted, resulting in a shift in the peak position.
[0098] Example 11: As Figure 29 As shown, the joint / complete 3D search space is separated into two independent spaces, in which... and Use in the extended space To estimate by random guessing Compared to Example 5, the beam pointing pattern is not significantly distorted, only the power decreases, so the peak position remains unchanged.
[0099] Example 12: such as Figure 30 As shown, the joint / complete 3D search space is separated into two independent spaces, in which... and Use in the extended space To estimate by random guessing Compared to Example 6, the beam pointing pattern is significantly distorted, resulting in a shift in the peak position.
[0100] Example 13: such as Figure 31 As shown, the joint / complete 3D search space is separated into two independent spaces, in which... and Use in the extended space To estimate by random guessing Compared to Example 7, the beam pointing pattern has been significantly distorted, resulting in a shift in the peak position.
[0101] Next, in order to solve practical problems, such as Figure 32 , Figure 33 and Figure 34 As shown, the detection priority of peak position detection in each search space is determined by the spatial selectivity (i.e., spatial degrees of freedom) of the received beamforming output power within the corresponding space. Specifically, search spaces with higher spatial selectivity are assigned higher detection priority and their peak position detection should be performed first. For geostationary orbit systems, along the search direction... (or search space) )right The detection of [certain types] usually has the lowest detection priority. This is because... For example, in the application scenario of geostationary orbit systems, , .
[0102] Finally, in order to solve practical problems, such as Figure 35 As shown, the present invention is aimed at the first... Search directions Assuming a temporary random guess estimate , The probability of a random guess failing should be as low as possible; and The ratio of the guessing blind zone (or dead zone) (i.e., the area where the received beam output power is extremely low and communication is impossible) to the entire search space should also be as low as possible. First, a temporary random guess can be used to approximate the peak position along a spatially flat (i.e., decoupled) search direction. Then, in the reduced-dimensional search space, peak position detection is performed using the aforementioned temporary random guess, which results in a slight distortion of the overall received beamforming output power curve (only a slight decrease in power level) (see...). Figure 29 The performance degradation caused by the aforementioned random guessing in the guessing blind zone is limited; for example, a 20% power reduction occurs with approximately a 3% probability. Next, the system checks the received beamforming output power. Does it meet the following minimum power requirement conditions: : .
[0103] if The system will then set another value with a specific angular displacement based on the previous estimate (random guess). The guesses were made and the search process was re-executed.
[0104] Appendix: Array Index Mapping Array index mapping: such as Figure 36 The diagram illustrates the conversion (mapping) between one-dimensional and three-dimensional indices. This mapping is based on the following relationship: definition We have a mapping from 3D indexes to 1D indexes. And the mapping from one-dimensional indexes to three-dimensional indexes. .
[0105] Example 1: Converting a 3D index to a 1D index: Given ,and , , Length of direction , , They are respectively Searching a one-dimensional index :
[0106] Example 2: Convert a one-dimensional index to a three-dimensional index: Given as well as Searching for 3D indexes :
[0107]
[0108]
[0109] Example 3: Converting a two-dimensional index to a one-dimensional index: Given , as well as Searching a one-dimensional index :
[0110] Example 4: Convert a one-dimensional index to a two-dimensional index: Given , as well as Searching two-dimensional index :
[0111]
[0112] This invention decouples the joint / complete three-dimensional search space into a low-dimensional search space through spatial decoupling, which significantly reduces the number of searches, reduces computational resource consumption and search time, and thus reduces search complexity.
[0113] Multi-stage peak position detection, combined with progressively smaller search steps, enables the system to complete peak positioning in fewer attempts, shortening the time required for the initial search and thus improving search efficiency.
[0114] This invention effectively avoids the accumulation of errors from single-space detection and improves the accuracy of the final peak position by interactively detecting the peak position between the dimensionality-reduced search space and the decoupled search space and finally reconstructing the result.
[0115] This invention reduces the probability of search failure when the peak value falls into the low-power region by using angle jump design and random guessing mechanism, thereby improving system robustness and avoiding the power dead zone problem.
[0116] This invention is particularly suitable for the initial search of geosynchronous orbit systems and is applicable to dynamic communication scenarios. It can support moving user terminals to quickly and stably establish an initial link with the target satellite during the cold start phase.
Claims
1. A space decoupling method for preliminary search of a geosynchronous orbit system, characterized in that, This includes the following steps performed sequentially: Step 1: Analyze the spatial selectivity of the received beamforming output and decouple at least one one-dimensional decoupled search space from the joint / complete three-dimensional search space; Step 2: Based on the peak position guess from the decoupled search space, establish and form a corresponding two-dimensional reduced search space, and perform the first peak position detection in the reduced search space; Step 3: Based on the detection results of the first peak position detection, perform the second peak position detection in the decoupled search space; as well as Step 4: Integrate the peak position detection results from the decoupled search space and the dimensionality reduction search space to reconstruct the final peak position.
2. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, The decoupling described in step one is based on spatial selective analysis of the beam pointing pattern on an analog mechanical angle scale to obtain the corresponding spatial degrees of freedom.
3. The space decoupling method for preliminary search of geosynchronous orbit systems as described in claim 1, characterized in that, The dimensionality reduction search space corresponds to the equivalent dimensionality reduction receiver beamforming output power obtained by pre-specifying some variables in the joint objective function corresponding to the joint / complete three-dimensional search space.
4. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, In step three, temporary random guesses or pre-specified estimates are used for the search parameters of other non-detection directions, and the peak position is re-detected in the decoupled search space.
5. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, Steps two and three are completed through a multi-stage peak position detection procedure. Multi-stage detection includes at least two search stages of different lengths to gradually narrow the search range and improve detection accuracy.
6. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, Angle jumps are performed during the detection of the second peak position in the decoupled search space to avoid power dead zones in the detection of the first or second peak position.
7. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 5, characterized in that, The search step size is set in a decreasing manner, so that a larger step size is used in the initial stage and a smaller step size is used in the subsequent stages.
8. The space decoupling method for preliminary search of geosynchronous orbit systems as described in claim 1, characterized in that... Further, after completing step four, it is confirmed whether the received beamforming output power meets the minimum required power condition. If the received beamforming output power is less than or equal to the minimum required power condition, another guess with a specific angular displacement is set based on the peak position guess of the decoupled search space, and steps two, three and four are executed again in sequence.
9. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, During the peak position detection process in the reduced-dimensional search space and the decoupled search space, the peak position detection is performed according to the detection priority order determined by the spatial selectivity of the received beamforming output power of each search space. Those with higher spatial selectivity are assigned higher detection priority.
10. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, The joint / complete 3D search space is decoupled into two independent search spaces, including a joint dimensionality reduction search space consisting of two search directions in the joint / complete 3D search space, and a decoupled search space consisting of the other search direction in the joint / complete 3D search space.
11. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, The decoupled search space has lower spatial selectivity compared to the dimensionality-reduced search space.
12. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, Dimensional reduction search space has higher spatial selectivity compared to decoupled search space.
13. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, The decoupled search space is composed of a single search direction in the joint / complete three-dimensional search space.
14. The space decoupling method for preliminary search of a geosynchronous orbit system as described in claim 1, characterized in that, The dimensionality reduction search space is composed of complex search directions in the joint / complete three-dimensional search space.