Array navigation dual-gain precision observation method and device

Abstract

The application provides an array navigation double-gain precise observation method and device, and relates to the technical field of satellite navigation array signal processing. In the application, first, coherent synthetic observation and spatial flow observation are generated based on multi-array element synchronization samples, so that the carrier-to-noise ratio gain and spatial response difference information are utilized; second, a joint covariance matrix containing cross-correlation items of the two types of observations is constructed, the statistical correlation between homologous observations is explicitly processed, and the relevant information is avoided from being repeatedly counted; third, combined with the joint observation model, projection or Schur complement is used to eliminate continuous interference parameters, effective condition information reflecting the actual integer discrimination ability of each candidate configuration is obtained; finally, the discrimination distance of the dangerous integer direction is calculated according to the condition information, and the target configuration is selected by comprehensively outputting the signal-to-noise ratio, the phase fidelity index and the satellite response amplitude, so that the reliability of the integer ambiguity fixing and the navigation solution is improved.

Description

Technical Field

[0001] This invention relates to the field of satellite navigation array signal processing technology, and specifically to a dual-gain precision observation method and apparatus for array navigation. Background Technology

[0002] Satellite navigation receivers can employ multiple antenna elements to simultaneously receive satellite signals and improve the reception quality through beamforming, interference nulling, or other space processing. Coherent combining typically compresses multi-element data into a single output to improve the effective carrier-to-noise ratio and carrier tracking sensitivity. During multi-element reception, relative response differences between elements also arise, depending on array geometry, satellite orientation, carrier attitude, and space propagation state.

[0003] In related technologies, coherent synthesized outputs and other spatial response outputs may be formed from the same batch of multi-element synchronized samples through different digital processing. Since each output shares the original sample noise, their noise may exhibit cross-correlation. In some processing steps, the statistical correlation between different outputs is represented using independent or simplified models. Continuous interference parameters such as common complex gain, channel amplitude and phase error, array attitude error, phase center deviation, multipath propagation, non-line-of-sight, and interference parameters may also be coupled with navigation state and integer ambiguities.

[0004] Furthermore, the evaluation of array configurations typically employs metrics such as output signal-to-noise ratio, average gain, or average information content. When observation correlations, continuous parameter coupling, or local distributions of integer candidates are complex, these metrics may deviate in characterizing the actual observation quality and the resolvability of integer candidates for different configurations, thereby affecting the results of integer ambiguity fixing and navigation state resolution. Summary of the Invention

[0005] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a dual-gain precision observation method and apparatus for array navigation, which solves the technical problems of how to accurately handle the cross-correlation between different observations formed by the same multi-element synchronous samples, eliminate the coupling between continuous interference parameters and integer ambiguity, and select array configuration by combining the ability to discriminate dangerous integer directions.

[0006] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides an array navigation dual-gain precision observation method, comprising: Acquire multi-element synchronization samples and their noise or interference statistics from a satellite navigation receiver; Based on multi-element synchronous samples, coherent synthetic observations and spatial manifold observations are generated; wherein, the coherent synthetic observations are used to improve the carrier-to-noise ratio, and the spatial manifold observations preserve the spatial response differences between different array elements, different subarrays, different spatial configurations, or different physical observation states. Based on noise or interference statistics, a joint covariance matrix is ​​constructed, which includes cross-correlation terms of coherent synthetic observations and spatial manifold observations generated from the same multi-element synchronous samples. Select at least two of the following: carrier observation, code observation, coherent synthesis observation, and spatial manifold observation, and at least one of them must be coherent synthesis observation or spatial manifold observation, and construct a joint observation model that includes continuous navigation state, integer ambiguity, and continuous interference parameters. Different spatial processing configurations or physical observation configurations are used as candidate configurations. For each candidate configuration, the observation noise covariance matrix corresponding to the selected observation is determined from the joint covariance matrix. Combined with the joint observation model, the continuous interference parameters are eliminated by projection or Schur complement to obtain effective condition information corresponding to each candidate configuration and related to the integer ambiguity. Using valid condition information, calculate the discrimination distance of each candidate configuration in multiple dangerous integer directions, and determine the target configuration from the candidate configurations based on at least two of the following: output signal-to-noise ratio, phase fidelity index, satellite response amplitude, and minimum discrimination distance; Based on the target configuration, floating-point estimation, integer least squares estimation, and integrity checks are performed to obtain the navigation state solution results.

[0007] In a second aspect, the present invention provides an array-guided dual-gain precision observation device, comprising: The sample acquisition module is used to acquire multi-element synchronization samples and their noise or interference statistics from the satellite navigation receiver. The observation generation module is used to generate coherent composite observations and spatial manifold observations based on multi-element synchronous samples; wherein, the coherent composite observations are used to improve the carrier-to-noise ratio, and the spatial manifold observations preserve the spatial response differences between different array elements, different subarrays, different spatial configurations, or different physical observation states; The covariance construction module is used to construct a joint covariance matrix based on noise or interference statistics, including cross-correlation terms of coherent synthetic observations and spatial manifold observations generated from the same multi-element synchronous samples. The joint modeling module is used to select at least two of the following: carrier observation, code observation, coherent synthesis observation, and spatial manifold observation, with at least one of them being coherent synthesis observation or spatial manifold observation, to construct a joint observation model that includes continuous navigation state, integer ambiguity, and continuous interference parameters. The information determination module is used to select different spatial processing configurations or physical observation configurations as candidate configurations, and for each candidate configuration, to determine the observation noise covariance matrix corresponding to the selected observation from the joint covariance matrix. Combined with the joint observation model, the continuous interference parameters are eliminated by projection or Schur complement to obtain effective condition information corresponding to each candidate configuration and related to the integer ambiguity. The configuration selection module is used to calculate the discrimination distance of each candidate configuration in multiple dangerous integer directions using effective condition information, and to determine the target configuration from the candidate configurations based on at least two of the following: output signal-to-noise ratio, phase fidelity index, satellite response amplitude, and minimum discrimination distance. The navigation solution module is used to perform floating-point estimation, integer least squares estimation and integrity checks based on the target configuration to obtain the navigation state solution results.

[0008] As a further improvement to the above-mentioned array navigation dual-gain precision observation method or device, the generation process of the coherent synthetic observation includes: The coherent synthesis weights are determined based on at least one of the following criteria: maximum signal-to-noise ratio criterion, minimum variance distortionless response criterion, linear constraint criterion, multi-star common weight constraint, and interference suppression constraint. Based on the coherent synthesis weights, the multi-element synchronous samples are coherently synthesized to generate the coherent synthesized observations.

[0009] As a further improvement to the above-mentioned array navigation dual-gain precision observation method or device, the generation process of the space manifold observation includes: Based on the multi-element synchronization samples, at least one of the following is obtained: subarray output, amplitude ratio between elements, phase difference between elements, complex projective coordinates after eliminating common complex gain, and phase transfer between different spatial processing configurations, and is determined as the spatial manifold observation.

[0010] As a further improvement to the aforementioned array navigation dual-gain precision observation method or device, the joint covariance matrix is ​​constructed as follows: When the coherent composite observation and the spatial manifold observation are generated from the same multi-element synchronous sample, the coherent composite weights used to generate the coherent composite observation and the spatial manifold processing matrix used to generate the spatial manifold observation are determined. Based on the original array element noise or interference covariance matrix, the coherent synthesis weights, and the spatial manifold processing matrix, the covariance of the coherent synthesis observation, the covariance of the spatial manifold observation, and the cross covariance of the two are determined, and the joint covariance matrix is ​​formed accordingly.

[0011] As a further improvement to the above-mentioned array navigation dual-gain precision observation method or device, the continuous interference parameters include at least one of common complex gain, channel amplitude and phase parameters, multipath parameters, interference direction, and interference intensity; the process for eliminating the continuous interference parameters includes: Based on the observation noise covariance matrix and the joint observation model, determine the tangent space or information matrix block corresponding to the continuous interference parameters; The joint observation model is projected onto the orthogonal complement space of the tangent space corresponding to the continuous interference parameters, or the Schur complement is obtained by dividing the information matrix corresponding to the continuous interference parameters into elimination blocks, so as to obtain effective conditional information related to the integer ambiguity.

[0012] As a further improvement to the above-mentioned array navigation dual-gain precision observation method or device, the process for determining the multiple dangerous integer directions includes: The plurality of dangerous integer directions are determined based on at least one of integer least squares candidate differences, integer Voronoi neighbors, short integer vector sets, and historical error fixation events.

[0013] As a further improvement to the above-mentioned array navigation dual-gain precision observation method or device, performing the floating-point estimation includes: The floating-point ambiguity is updated using the spatial manifold observations and the covariance and cross-correlation terms in the observation noise covariance matrix corresponding to the spatial manifold observations. Performing the integer least squares estimation includes: Based on the updated floating-point ambiguity, integer candidate scoring items are constructed or integer candidates are reordered using the spatial manifold observations.

[0014] Thirdly, the present invention provides a storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the array navigation dual-gain precision observation method as described in the first aspect.

[0015] Fourthly, the present invention provides an electronic device, the electronic device comprising: A processor, a memory, and program instructions stored in the memory and executable by the processor, wherein the processor, when executing the program instructions, implements the array navigation dual-gain precision observation method as described in the first aspect.

[0016] (III) Beneficial Effects This invention provides a method and apparatus for dual-gain precision observation of array navigation. Compared with the prior art, it has the following advantages: In this invention, firstly, coherent synthetic observations and spatial manifold observations are generated based on multi-element synchronous samples, enabling the utilization of carrier-to-noise ratio gain and spatial response difference information. Secondly, by constructing a joint covariance matrix containing cross-correlation terms of the two types of observations, the statistical correlation between observations from the same source is explicitly processed, avoiding the duplication of relevant information. Thirdly, combined with the joint observation model, projection or Shur complement is used to eliminate continuous interference parameters, obtaining effective conditional information reflecting the actual integer discrimination capability of each candidate configuration. Finally, the discrimination distance of dangerous integer directions is calculated based on this, and the target configuration is selected by comprehensively outputting the signal-to-noise ratio, phase fidelity index, and satellite response amplitude, thereby improving the reliability of integer ambiguity fixing and navigation solution. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a basic flowchart of a dual-gain precision observation method for array navigation provided in an embodiment of the present invention; Figure 2 A block diagram of an array navigation dual-gain precision observation method provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of carrier-to-noise ratio gain propagation and observation weight adjustment for coherent composite observation provided in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the elimination of continuous interference parameters and determination of effective condition information for a joint observation model provided in an embodiment of the present invention. Figure 5 A schematic diagram of switching and phase compensation of a space processing configuration or physical observation configuration provided in an embodiment of the present invention; Figure 6 This is a schematic diagram illustrating the integrity verification of a fixed integer result and a navigation state calculation result, provided as an embodiment of the present invention. Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0019] Component labeling explanation: 100 - electronic device, 101 - memory, 102 - processor, 103 - display. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] In the joint observation and configuration selection process of array navigation, it is necessary to accurately characterize the cross-correlation between multiple observations generated from the same multi-element synchronous sample, and to handle the coupling between continuous interference parameters such as channel amplitude and phase, multipath and integer ambiguity; at the same time, it is also necessary to enable the configuration evaluation to reflect the actual discrimination capability of multiple dangerous integer directions.

[0022] Furthermore, coherent composite observations primarily provide carrier-to-noise ratio gain, while spatial manifold observations mainly preserve the spatial response differences between different array elements, subarrays, spatial configurations, or physical observation states. Therefore, it is also necessary to collaboratively utilize both types of array observations based on explicitly handling the correlation of observations from the same source, and to consider factors such as output signal-to-noise ratio, phase fidelity index, satellite response amplitude, and minimum discrimination distance in the configuration selection.

[0023] To solve the above technical problems, see [link to relevant documentation]. Figure 1 The basic workflow of this invention is as follows: First, obtain the multi-element synchronization samples of the satellite navigation receiver and their noise or interference statistics.

[0024] Secondly, coherent synthetic observations and spatial manifold observations are generated based on the multi-element synchronous samples, and a joint covariance matrix including cross-correlation terms of the two types of observations from the same source is constructed.

[0025] Next, at least two of the following are selected: carrier observation, code observation, coherent synthesis observation, and spatial manifold observation, with at least one being coherent synthesis observation or spatial manifold observation, to construct a joint observation model; and using different spatial processing configurations or physical observation configurations as candidate configurations, for each candidate configuration, continuous interference parameters are eliminated by projection or Schur complement to obtain effective conditional information related to integer ambiguity.

[0026] Finally, the discrimination distance of each candidate configuration in multiple dangerous integer directions is calculated, and the target configuration is determined based on at least two of the output signal-to-noise ratio, phase fidelity index, satellite response amplitude and minimum discrimination distance. Based on the target configuration, floating-point estimation, integer least squares estimation and integrity check are performed to obtain the navigation state solution results.

[0027] like Figure 2 As shown, this embodiment provides a dual-gain precision observation method for array navigation, including: S101. Obtain multi-element synchronization samples and their noise or interference statistics from the satellite navigation receiver; S102. Based on multi-element synchronous samples, generate coherent composite observations and spatial manifold observations; wherein, the coherent composite observations are used to improve the carrier-to-noise ratio, and the spatial manifold observations preserve the spatial response differences between different array elements, different subarrays, different spatial configurations, or different physical observation states. S103. Based on noise or interference statistics, construct a joint covariance matrix including cross-correlation terms of coherent synthetic observations and spatial manifold observations generated from the same multi-element synchronous samples; S104. Select at least two of the following: carrier observation, code observation, coherent synthesis observation, and spatial manifold observation, and at least one of them is coherent synthesis observation or spatial manifold observation, and construct a joint observation model that includes continuous navigation state, integer ambiguity and continuous interference parameters. S105. Using different spatial processing configurations or physical observation configurations as candidate configurations, and for each candidate configuration, determining the observation noise covariance matrix corresponding to the selected observation from the joint covariance matrix, and combining the joint observation model, eliminating the continuous interference parameters through projection or Schur complement to obtain effective condition information corresponding to each candidate configuration and related to the integer ambiguity. S106. Using the effective condition information, calculate the discrimination distance of each candidate configuration in multiple dangerous integer directions, and determine the target configuration from the candidate configurations based on at least two of the following: output signal-to-noise ratio, phase fidelity index, satellite response amplitude, and minimum discrimination distance. S107. Based on the target configuration, perform floating-point estimation, integer least squares estimation and integrity check to obtain the navigation state solution results.

[0028] The "dual gain" referred to in this embodiment refers to two types of gains that complement each other but have different effects. Coherent composite observation primarily reflects sensitivity gain, i.e., improving the output signal-to-noise ratio or carrier-to-noise ratio through coherent superposition; spatial manifold observation primarily reflects conditional information gain, i.e., improving the distinguishability of navigation states or integer candidates by preserving the relative response differences between spatial channels. Conditional information gain is not equivalent to simply increasing the number of digital outputs; the multiple outputs obtained from the deterministic transformation of the same original sample (multi-element synchronized sample) are still constrained by the effective rank and joint statistical correlation of the original sample.

[0029] For the observation combination, carrier observation can refer to carrier phase observation formed by the receiver based on carrier tracking results, and code observation can refer to pseudorange or code phase observation formed based on code tracking results; coherent synthesis observation and spatial manifold observation are two types of array observations formed by array samples. Step S104, "selecting observations," determines the set of observations to be included in this joint estimation from the above observation types. Steps S105 and S106, "candidate configuration," determine how each observation is formed or under what physical state it is acquired; these correspond to observation type selection and observation configuration selection, respectively.

[0030] The following establishes a unified convention for the parameter symbols used in the formulas of this embodiment: Indicates the number of array element channels. Represents the number of outputs of the space manifold. Indicates that it is obtained from the same observation epoch or related integration interval. Multidimensional matrix element complex sample vector; Represents the continuous navigation state vector. Indicates integer ambiguity. This represents the continuous interference parameter.

[0031] This represents a candidate spatial processing configuration or a candidate physical observation configuration. Represents the set of candidate configurations. Indicates the current or previous configuration. This represents the target configuration determined from the set of candidate configurations. Indicates configuration The following are used to form coherent composite observations 3D complex weight vector, Indicates configuration The following is used to form Road space manifold observation Multidimensional complex processing matrix, and These represent coherent composite observations and spatial manifold observations, respectively.

[0032] Indicates configuration Next satellite The array of complex response vectors, Indicates satellite The power of the signal in the element domain, This indicates the output signal-to-noise ratio of the coherent composite observation; This represents the noise or interference covariance matrix in a multi-element sample. This represents the joint covariance matrix of coherent composite observations and spatial manifold observations. This represents the observation noise covariance matrix determined from the joint covariance matrix and corresponding to the selected observation. , and These represent the covariance of coherent composite observations, the cross covariance of coherent composite observations and spatial manifold observations, and the covariance of spatial manifold observations, respectively. The noise covariance symbol will not be used separately thereafter, or the unsubscripted covariance symbol will be used to represent the above statistics.

[0033] Indicates configuration The following is a joint observation vector formed by stacking the selected observations. This represents the corresponding observation noise vector. This represents the joint observation function. Indicates the joint observation function for integer ambiguity. Jacobian matrix; Let the Jacobian matrix represent the continuous parameters to be conditionally transformed, wherein the continuous parameters to be conditionally transformed include at least... In evaluating integer fuzziness When considering independent discrimination capability, it may also include the ability to distinguish integer ambiguity. Coupled and requiring conditional continuous navigation states . This represents the precision-weighted Jacobian matrix ratio of the continuous parameters to be conditionalized. This represents the continuous parameter information matrix to be conditionally quantified. This represents the weighted orthogonal complement matrix after subtracting the contribution of continuous parameter information to be conditionalized, weighted by the observation noise covariance. Indicates configuration Lower integer ambiguity The relevant valid condition information matrix.

[0034] Represents a non-zero integer ambiguity difference vector. Represents the set of dangerous integer directions. Indicates configuration In direction The discriminant distance on Indicates configuration The minimum discriminant distance in the set of dangerous integer directions. This represents the evaluation metric obtained by normalizing the output signal-to-noise ratio. This indicates a switch from the current configuration to the candidate configuration. The cost of switching , and This indicates the non-negative weight of the corresponding evaluation item. This represents the overall configuration evaluation value comprised of the above evaluation items. (Superscript) Indicates orthogonal complement, superscript Indicates conjugate transpose, superscript Indicates transpose, superscript It indicates a false reversal.

[0035] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0036] In step S101, the multi-element synchronization samples of the satellite navigation receiver and their noise or interference statistics are obtained.

[0037] Satellite navigation receivers receive satellite navigation signals through multiple array element channels and form multi-element synchronization samples under a common or mutually synchronized time reference. For example, for receivers containing... A receiver with multiple array elements can arrange the complex correlation outputs within the same sampling time or the same correlation integration interval into an M-dimensional complex vector. Here, "synchronization" means that the samples of each channel can correspond to the same observation epoch or have completed a time alignment sufficient for coherent processing, and does not require that each RF channel has exactly the same amplitude and phase response.

[0038] To ensure comparability of coherent superposition and phase relationships between array elements, each element channel can share a reference clock, sampling clock, and local oscillator source, or they can be sampled separately and aligned in time, frequency, and phase based on calibration values. Furthermore, the receiver can store the correspondence between element numbers, sampling times, satellite numbers, frequency points, correlator branches, and physical array states, enabling the same batch of multi-element synchronous samples to simultaneously enter the coherent synthesis observation generation process and the spatial manifold observation generation process.

[0039] The noise or interference statistics may include at least one of the following: element domain noise covariance, interference plus noise covariance, channel variance, inter-channel cross-covariance, and statistical estimates of interference direction or intensity. For example, the statistics can be estimated using residual samples from satellite-free intervals, idle frequency bands, correlator non-peak branches, or the current observation window, and updated using a sliding window or recursive method based on environmental changes. The statistics can be estimated directly from the original element samples or corrected by combining them with receiver front-end calibration results.

[0040] Optionally, when the statistical sample size is small or the condition number of the covariance matrix is ​​large, diagonal loading or temporal smoothing can be performed based on the receiver noise level, sample length, or existing calibration results. The statistical information is then used to determine the coherent synthesis weights, construct the joint covariance, and evaluate the observation noise for different candidate configurations.

[0041] By retaining the common time index of the original array element samples and statistical information in step S101, it is possible to determine whether coherent composite observations and spatial manifold observations originate from the same batch of samples, and calculate cross-correlation terms accordingly. If the two types of observations come from different epochs, different frequencies, or different physical states, their correlation is determined according to the actual sampling relationship and spatiotemporal correlation, rather than being assumed to be independent.

[0042] In step S102, coherent composite observations and spatial manifold observations are generated based on multi-element synchronous samples; wherein, the coherent composite observations are used to improve the carrier-to-noise ratio, and the spatial manifold observations preserve the spatial response differences between different array elements, different subarrays, different spatial configurations, or different physical observation states.

[0043] In one processing path of this step, multi-element synchronization samples are processed. Apply configuration Corresponding weighted values Coherent superposition is then performed to form a coherent composite observation. Configuration The coherent synthesis observation and output signal-to-noise ratio under these conditions can be exemplarily expressed by equations (1a) and (1b): (1a) (1b) Coherent composite observations can be complex correlation outputs used for carrier loop, code loop, or joint navigation estimation, or they can be further obtained from the complex correlation outputs as carrier phase, code phase, amplitude, or carrier-to-noise ratio observations.

[0044] When carrier phase observations or code phase observations are further formed based on the coherent synthesized output, the observation type is determined based on its actual generation path and its data representation in the joint observation model. The same amount of data is not repeatedly counted in different observation types simply because it has both carrier and code attributes; only when they form different observations and their correlation has been explicitly handled in the observation noise covariance matrix can they be included in the joint observation model as different observations.

[0045] The "coherent composite observation" referred to in this embodiment refers to array observation formed by coherently superimposing the same batch of array element samples while maintaining complex phase relationships. Its difference from general single-antenna carrier observation lies in that this observation includes the response after array complex weighting; its difference from array detection quantities that only output beam power lies in that this observation can retain the complex phase used for precise carrier processing. Coherent combining typically compresses multi-element information into a smaller output, thus focusing on improving tracking sensitivity, but may average out the relative spatial response between array elements.

[0046] Furthermore, the coherent synthesis weights can be determined based on the maximum signal-to-noise ratio criterion, the minimum variance distortion-free response criterion, the linearly constrained minimum variance criterion, the multi-satellite common weight criterion, the interference suppression criterion, or a combination thereof. Under correlated noise conditions, the weights can be determined using the noise or interference covariance and satellite array response obtained in step S101; under weak interference conditions, equal-gain synthesis after preset phase calibration can also be used.

[0047] In some implementations, multiple satellites share a set or a class of commonly constrained weights to ensure that the composite observations from different satellites maintain the same common complex gain as much as possible. In this way, the common complex gain and common phase remain cancelable during inter-satellite differential analysis. In contrast, if each satellite is assigned a maximum signal-to-noise ratio weight, different satellites will introduce different complex gains; in this case, the amplitude and phase errors of the corresponding gains can be incorporated as continuous interference parameters into the joint observation model in subsequent step S104, or compensated for through calibration and phase transfer.

[0048] To balance interference suppression and precise carrier availability, the weight design can further set lower limits for satellite response amplitude, distortion-free constraints in the desired direction, null constraints in the interference direction, weight norm constraints, and radio frequency dynamic range constraints. Phase fidelity requirements do not mean that the weights must remain constant, but rather that the satellite-by-satellite phase change after the weights are applied is predictable and compensable, and its residual uncertainty is within acceptable limits.

[0049] For further implementation methods and effect evaluation methods, see Figure 3 Under the condition that the loop bandwidth, coherent integration time, receiver dynamic conditions, and other error conditions remain unchanged, and coherent synthesis meets the phase fidelity requirement, the output carrier-to-noise ratio (CNR) of coherent synthesis observations is improved. This can be reflected in the observation noise model as a reduction in carrier phase thermal noise variance, thereby reducing the risk of cycle slip caused by noise. Furthermore, in joint navigation estimation, the observation weights can be adjusted based on the actual output CNR and joint covariance of coherent synthesis observations, and the effective condition information for integer ambiguity can be determined based on the adjusted observation weights. When other conditions are the same and the effective condition information increases, this can be manifested as a decrease in floating-point ambiguity covariance and a reduction in the integer candidate search region at a given confidence level. Thus, the CNR gain obtained by coherent synthesis can be used as the basis for observation noise modeling and joint estimation weight determination, rather than just as a receiver indicator independent of the navigation estimation process.

[0050] In another processing path within this step, spatial manifold observations are formed by preserving the relative responses between array elements, subarrays, digital spatial configurations, or physical observation states. Configuration The spatial manifold processing matrix is ​​uniformly represented as follows: Observations of the corresponding linear space manifold It can be expressed as equation (2): (2) For nonlinear observations such as complex ratios, normalized quantities, or projective coordinates, multiple complex outputs can be generated first, and then the required observations can be obtained through ratio or normalization mapping.

[0051] The "spatial manifold observation" referred to in this embodiment is not a traditional direction-finding result aimed solely at estimating the signal's direction of arrival. Instead, it serves as an observation source in the joint navigation model, used to constrain continuous navigation states, integer ambiguities, or coupled continuous parameters to be conditionalized. Its main difference from coherent synthetic observation is that coherent synthetic observation tends to compress multi-element responses into a single high signal-to-noise ratio output, while spatial manifold observation preserves the relative differences between spatial channels to provide an additional discriminative dimension for candidate states or candidate integers.

[0052] For example, spatial manifold observations may include at least one of the following: original element complex correlation quantities, multiple subarray outputs, amplitude ratios and phase differences between elements or subarrays, complex projective coordinates after eliminating common complex gains, response differences between different digital spatial configurations, phase transfer quantities before and after configuration switching, and observations under multiple non-redundant physical observation states. Here, spatial configuration may refer to digital configurations such as processing weights, subarray combinations, or null parameters, and physical observation states may refer to array attitude, physical subarrays, frequency points, observation times, or trajectory states.

[0053] The complex ratio between array elements can be expressed as the amplitude ratio and phase difference between array elements, or, after eliminating the common complex gain, as complex projective coordinates. Specifically, taking array element 1 as the reference array element, let... and They represent the first Given the complex samples or complex correlation outputs corresponding to the array element and the reference array element, the complex ratio between array elements can be expressed as: .in It can be decomposed into amplitude ratio and phase difference It can also be directly used as a complex observation.

[0054] In some implementations, a stable array element or subarray is selected as a reference, and the complex outputs of other array elements or subarrays are compared with the reference output to form complex projective coordinates that do not change with the common complex gain. Thus, the common propagation term from the satellite to the array, the receiver clock error, or the complex gain shared by each channel can be canceled out, and the remaining relative response is mainly related to the satellite orientation, array geometry, carrier attitude, channel relative amplitude and phase, and multipath structure. If the reference output amplitude is too low, the reference channel can be switched, or unity norm normalization and projected coordinates can be used to avoid numerical instability caused by direct comparison.

[0055] Furthermore, spatial manifold observations can be generated across multiple satellites, frequencies, and epochs, and transformed to a unified coordinate system based on the array calibration model. For observations before and after configuration switching, the phase transfer can be predicted using known processing weights or array phase center displacement; residual quantities that cannot be fully predicted are treated as continuous interference parameters or additional observation noise.

[0056] It is important to note that spatial manifold observations do not automatically add independent information. If multiple spatial manifold outputs and coherent composite outputs are all from the same batch... The array element samples are generated deterministically, and their effective independent dimension is limited by the batch of array element samples and their noise structure. Only when the physical states such as array attitude, frequency, observation time, or spatial location change and new actual observations are formed can new independent information be obtained under the corresponding spatiotemporal correlation conditions. This statistical boundary is reflected in the subsequent step S103 through joint covariance and effective rank processing.

[0057] In step S103, based on noise or interference statistics, a joint covariance matrix is ​​constructed, which includes cross-correlation terms of coherent synthetic observations and spatial manifold observations generated from the same multi-element synchronous samples.

[0058] Based on the noise or interference statistics obtained in step S101, this step constructs the joint covariance matrix of coherent synthetic observations and spatial manifold observations. When both types of observations are generated from the same batch of multi-element synchronous samples, they typically share element noise, residual interference, and front-end errors, and their cross-correlation is generally not zero. and coherent synthesis weights and spatial manifold processing matrix The joint covariance matrix can be determined according to equations (3a) to (3d). and its covariance block: (3a) (3b) (3c) (3d) in, and its conjugate transpose Located in off-diagonal blocks, used to represent cross-correlation between two types of observations.

[0059] The "joint covariance matrix" referred to in this embodiment includes at least the covariance of the coherent synthetic observations, the covariance of the spatial manifold observations, and the cross-covariance between the two. Explicitly handling cross-correlation means preserving off-diagonal correlations in covariance representation, whitening, information matrix construction, or equivalent statistical processing, rather than assuming that two observations generated from the same source are independent. This preserves the complementary response characteristics of the two observations while ensuring that their shared noise is included in subsequent estimation only according to its actual statistical contribution.

[0060] For observations of nonlinear spatial manifolds, such as element ratios and normalized projective coordinates, a first-order expansion of the observation map can be performed at the current operating point. The covariance of the spatial manifold observations can be propagated using its Jacobian matrix, and simultaneously, its cross-covariance with coherent synthetic observations can be propagated. Alternatively, the corresponding sample covariance can be directly estimated based on synchronous samples. The method employed should be adapted to the available sample size and degree of nonlinearity at the receiver.

[0061] As a further implementation, when the dimension of the joint output is greater than the independent dimension that the original array element samples can support, or when multiple processing vectors are approximately collinear, the joint covariance matrix may be singular or near-singular. In this case, eigenvalue decomposition or singular value decomposition can be performed on the joint covariance, the effective rank can be determined based on the noise floor and numerical precision, and whitening can be performed in the effective subspace; alternatively, a pseudo-inverse can be used, or redundant outputs with small marginal information increments can be deleted.

[0062] Statistically speaking, deterministic processing of the same batch of original array element samples (multi-element synchronous samples) will not unconditionally generate more information than the original samples. By combining covariance, effective rank, and pseudo-inverse processing, the rank of subsequent effective conditional information can be ensured not to exceed the independent dimension supported by the original observations. As an equivalent implementation, the original array element samples can also be used directly as sufficient statistics to build the model, and coherent synthesis and spatial manifold processing can be used to construct the model matrix or evaluation index.

[0063] Furthermore, the joint covariance matrix can be arranged according to the available observations among carrier observations, code observations, coherent synthesis observations, and spatial manifold observations, and at least includes a cross-correlation term between coherent synthesis observations and spatial manifold observations generated from the same multi-element synchronous samples. When only a portion of the observations are selected in subsequent step S104, in step S105, based on the index corresponding to the selected observations, the corresponding rows and columns are extracted from the joint covariance matrix to form a covariance submatrix including the covariance of each selected observation and its mutual cross-covariance, and the covariance submatrix is ​​determined as the observation noise covariance matrix. When the selected observations include carrier observations or code observations, their covariance and the cross-covariance between them and other selected observations can be determined based on the actual generation relationship of the carrier observations or code observations, and the corresponding covariance terms are included in the joint covariance matrix; the corresponding cross-covariance is set to zero only when the receiver structure, sample source, and time relationship support the statistical independence of the corresponding observations.

[0064] In step S104, at least two of the following are selected: carrier observation, code observation, coherent synthesis observation, and spatial manifold observation, and at least one of them is coherent synthesis observation or spatial manifold observation, to construct a joint observation model including continuous navigation state, integer ambiguity and continuous interference parameters. At least two observations are selected from carrier observations, code observations, coherent combined observations, and spatial manifold observations, with at least one of the selected observations being either a coherent combined observation or a spatial manifold observation. The "at least two observations" refers to at least two actually different observations, not different attribute names of the same observation; "actually different" does not imply statistical independence, and the correlation between observations is handled according to steps S103 and S105. For example, the selected observations may include carrier observations and coherent combined observations, carrier observations and spatial manifold observations, code observations and coherent combined observations, code observations and spatial manifold observations, coherent combined observations and spatial manifold observations, or three or four of the four types of observations. For the selected observation combination, in step S105, the corresponding observation noise covariance matrix is ​​determined based on the joint covariance matrix constructed in step S103, ensuring that the dimension and covariance terms of the observation noise covariance matrix match the actual observations entering the joint observation model.

[0065] After arranging the selected observations by satellite, frequency, epoch, or observation type, the joint observation model shown in equation (4) can be constructed.

[0066] (4) in, For configuration Joint observation vectors under; continuous navigation state It may include one or more of the following: position, velocity, clock bias, baseline, attitude, and rate of change; integer ambiguity. Integer unknowns in corresponding carrier observations; continuous interference parameters For continuous values ​​and continuous navigation state or integer fuzziness Parameters that are coupled but are not currently the primary navigation output object; This represents the corresponding observation noise.

[0067] The "continuous interference parameter" referred to in this embodiment can also be called the continuous nuisance parameter. It is not limited to the physical parameters of external radio frequency interference, but can include at least one of the following: common complex gain, amplitude and phase parameters of each channel, array attitude error, phase center deviation, multipath parameter, non-line-of-sight parameter, interference direction, interference intensity, per-satellite amplitude and phase deviation introduced by processing weights, and time alignment deviation. These parameters are explicitly included in the model to show that they will consume the interpretability of observations for integer ambiguities or navigation states.

[0068] Furthermore, it is understandable that the difference between continuous disturbance parameters and observation noise lies in the fact that the former is modeled as an unknown continuous quantity within the current estimation window and can be jointly estimated, projected away, or constrained by priors; the latter represents the random residuals that still exist after a given state and parameters, and is described by the observation noise covariance. The difference between continuous disturbance parameters and integer ambiguities lies in the fact that the former takes values ​​in a continuous domain, while the latter is constrained by integers. For non-target quantities that change randomly over time, they can also be modeled in state form, and prior information can be provided based on their dynamic model.

[0069] For example, the carrier and code observation components can employ conventional pseudorange, carrier phase, or differential observation equations; the coherent synthesis component includes the satellite complex response and phase shift after weighting; and the spatial manifold component includes the mapping between array geometry, satellite orientation, carrier attitude, and channel relative amplitude and phase. Furthermore, the joint model can be linearized near the current floating-point state to obtain the Jacobian matrices corresponding to integer ambiguities, continuous navigation states, and continuous interference parameters.

[0070] To make the model identifiable, the common phase, reference array elements, reference satellite, or reference epoch can be normalized, and prior constraints can be added to the calibrated channel parameters, attitude, or phase center parameters that have a reliable error range. The normalization or priors are used to eliminate inherent degrees of freedom or utilize known calibration information; unknown continuous disturbance parameters should not be directly treated as known constants.

[0071] It should be noted that, in order to simultaneously utilize the sensitivity gain provided by coherent synthetic observations and the spatial response differences preserved by spatial manifold observations, in a preferred embodiment, this step includes both the coherent synthetic observations and the spatial manifold observations in the joint observation model. In this case, the joint observation model includes at least coherent synthetic observations and spatial manifold observations, and may further include at least one of carrier observations and code observations.

[0072] Furthermore, when coherent composite observations and spatial manifold observations are generated from the same multi-element synchronous samples, subsequent step S105 selects the covariance terms and their cross-correlation terms corresponding to the two types of array observations from the joint covariance matrix to form the observation noise covariance matrix used in the joint observation model. Thus, coherent composite observations and spatial manifold observations participate in parameter estimation as interrelated observations, rather than being treated as statistically independent observations.

[0073] In the preferred embodiment described above, coherent combined observations are primarily used to provide carrier-to-noise ratio gain, while spatial manifold observations are primarily used to provide relative response information between array elements, subarrays, spatial configurations, or physical observation states. The joint observation model incorporates both types of array observations, which is beneficial for retaining sensitivity and spatial condition information while explicitly handling cross-correlation from the same source. Accordingly, subsequent explanations regarding continuous interference parameter elimination, determination of effective condition information, and discrimination of dangerous integer directions can be preferentially developed based on this embodiment.

[0074] In step S105, different spatial processing configurations or physical observation configurations are used as candidate configurations. For each candidate configuration, the observation noise covariance matrix corresponding to the selected observation is determined from the joint covariance matrix. Combined with the joint observation model, the continuous interference parameters are eliminated by projection or Schur complement to obtain effective condition information corresponding to each candidate configuration and related to the integer ambiguity.

[0075] The aforementioned "candidate configuration" refers to a member to be evaluated in the configuration set. For example, a space processing configuration may include different coherent synthesis weights, manifold processing matrices, subarray combinations, reference elements, null directions, null depths, or whitening methods; a physical observation configuration may include different array attitudes, physical subarrays, frequencies, observation times, or carrier trajectory states. The candidate set may contain only space processing configurations, only physical observation configurations, or both types of configurations; each candidate has clearly defined processing parameters or physical states.

[0076] Candidate configurations can originate from pre-established weight or physical state libraries, online optimization results under the current covariance and satellite geometry, discrete combinations satisfying hardware constraints, or neighborhood configurations generated around the current configuration. Thus, each candidate configuration corresponds to a defined observation generation relationship. When the spatial processing matrix changes, the covariance transformation and cross-correlation terms in step S103 change accordingly; when the physical observation state changes, the satellite array response, observation model, and actual spatiotemporal correlation change accordingly.

[0077] For each candidate configuration, rows and columns corresponding to the observations selected in step S104 are extracted from the joint covariance matrix to obtain a covariance submatrix including the covariance of each selected observation and their cross covariances. When the selected observations have undergone differencing or other linear transformations, the covariance submatrix is ​​transformed based on the same transformation relationship; the covariance submatrix without the above transformation or the covariance matrix after the above transformation is determined as the corresponding observation noise covariance matrix. Subsequently, the processing weights, spatial manifold, satellite response, or physical state under the candidate configuration are substituted into the joint observation model. If the same candidate configuration only changes some observation generation relationships, the corresponding statistics can be used for the remaining observations; if the configuration change changes the sample source, reference channel, or frequency, the corresponding covariance terms and cross covariance terms are rearranged or updated.

[0078] The "effective condition information" referred to in this embodiment refers to the equivalent information that can still be used for integer ambiguity after projecting or Schur complementing the continuous interference parameters when the continuous interference parameters are unknown and coupled with integer ambiguity. It is different from directly reading the diagonal block of the integer information matrix, which is equivalent to treating the continuous interference parameters as known and may overestimate the integer discrimination ability; the effective condition information deducts the information used to interpret or estimate the continuous interference parameters.

[0079] It should be noted that the elimination of the continuous parameters to be conditionalized by projection or Shur complement in step S105 is to determine the effective condition information corresponding to the candidate configuration, and does not mean that the corresponding parameters are deleted from the subsequent navigation estimation model; in the subsequent step S107, the continuous interference parameters and continuous navigation state can still be jointly estimated based on the target configuration.

[0080] Continuing with the preferred implementation method described above, both coherent synthetic observations and spatial manifold observations are selected into the joint observation model, and carrier observations and / or code observations can be further selected.

[0081] Figure 4 The following example illustrates the scenario where both coherent composite observations and spatial manifold observations are included in the joint observation model. Figure 4As shown, for complex observations in the joint observation vector, the real and imaginary parts are expanded separately to form a real-valued joint observation vector. The corresponding observation noise covariance matrix is ​​then transformed based on the same expansion relation. The transformed observation noise covariance matrix is ​​still denoted as... The real-valued joint observation model is linearized and based on... After weighting, the integer fuzzyness Jacobian matrix can be... With the continuous parameter Jacobian matrix to be conditionalized Block partitioning. When using tangent space projection, the joint observation model is projected based on the orthogonal complement space of the tangent space corresponding to the continuous disturbance parameters; when using Schur complement, the Schur complement is obtained by partitioning the information matrix corresponding to the continuous disturbance parameters into blocks as elimination blocks.

[0082] Specifically, the precision-weighted Jacobian matrix can be determined sequentially according to equations (5a) to (5c). Continuous parameter information matrix to be conditionally categorized and weighted orthogonal complement matrix And obtain the effective conditional information matrix of integer ambiguity according to equation (5d). All the following expressions are based on the real-valued representation and uniformly use the transpose notation. .

[0083] (5a) (5b) (5c) (5d) It should be noted that equations (5a) to (5d) above apply to the non-singular observation noise covariance matrix obtained after effective rank whitening or redundant observation reduction. When retaining the singular matrix representation, the above operations can be performed within the effective eigenspace, or the equations can be modified accordingly. Consistent replacement The superscript in equation (5c) This represents the Moore-Penrose pseudoinverse representation, used to process continuous parameter information matrices to be conditionally categorized. The situation of being below rank.

[0084] Using projection or Shure time compensation, at least the continuous interference parameters should be eliminated. The impact on integer ambiguity information; in implementations that require evaluating the independent discrimination capability of integer ambiguity, the continuous navigation state coupled with integer ambiguity can also be considered. Conditionalization is performed simultaneously. When the corresponding matrix is ​​not of full rank, a pseudo-inverse is used; when the continuous parameter to be conditionalized has reliable prior information, the corresponding prior information can be incorporated into the information matrix of the continuous parameter to be conditionalized.

[0085] In step S106, using the valid condition information, the discrimination distance of each candidate configuration in multiple dangerous integer directions is calculated, and the target configuration is determined from the candidate configurations based on at least two of the output signal-to-noise ratio, phase fidelity index, satellite response amplitude, and minimum discrimination distance.

[0086] Using the valid condition information corresponding to each candidate configuration, this step calculates the discrimination distance of that candidate configuration in multiple dangerous integer directions. For non-zero integer ambiguity difference vectors... The given configuration can be calculated using equation (6). The smaller the discriminant distance, the smaller the difference. The two integer candidates are more likely to be confused under the current observation and parameter coupling conditions.

[0087] (6) For the dangerous integer direction set The minimum discriminant distance shown in equation (7) can be further taken. This serves as the worst-case direction indicator for the candidate configuration.

[0088] (7) in, This represents the minimum value operator.

[0089] The "minimum discrimination distance" referred to in this embodiment is the minimum discrimination distance among all discrimination distances for the same candidate configuration within the set of dangerous integer directions, i.e., the worst-case discrimination margin of the candidate configuration in the currently evaluated dangerous integer directions. The minimum discrimination distance is different from the unidirectional discrimination distance in a certain predetermined integer direction, and also different from the average information index obtained by averaging over all directions; the larger its value, the greater the degree of separation between the two most easily confused integer candidates in the current set of dangerous integer directions.

[0090] In this embodiment, the "dangerous integer direction" refers to the non-zero integer difference vector in the integer candidate space that is more likely to be erroneously fixed, rather than the spatial direction of arrival of the satellite signal. It can be derived from the difference between the best and second-best candidates obtained from the current integer least squares search, integer Voronoi neighbors, short integer vectors with small moduli, low-cost directions selected based on floating-point covariance, and directions recorded in historical erroneous fixation events. The set of dangerous integer directions can be updated as satellite geometry, observation quality, and floating-point solutions change.

[0091] Compared to average signal-to-noise ratio, trace, or average eigenvalue, minimum discriminant distance directly focuses on the most easily confused integer difference directions. In some implementations, the distance is first calculated or approximated from a larger set of integer directions, and then several directions with smaller distances are retained for real-time configuration evaluation; alternatively, dangerous integer directions from the previous epoch can be merged with the current search results to balance computational complexity and direction continuity.

[0092] The evaluation metrics for candidate configurations include at least two of the following: output signal-to-noise ratio (SNR), phase fidelity, satellite response amplitude, and minimum discrimination distance. For example, the output SNR can be converted into a dimensionless evaluation quantity. and form according to formula (8a) Minimum discriminant distance and switching costs The overall evaluation value is composed of ; , and The relative effects of each evaluation item are adjusted, and the target configuration is then determined according to equation (8b). Alternatively, constraint optimization or hierarchical screening methods can be used to first eliminate candidates that do not meet the phase fidelity or satellite response requirements, and then sort them according to the minimum discrimination distance and output signal-to-noise ratio.

[0093] (8a) (8b) in, This represents the independent variable that maximizes the objective function.

[0094] The "phase fidelity index" referred to in this embodiment can include two interrelated aspects. First, the complex response amplitude of the target satellite after space processing should not be too low to avoid the phase becoming unstable when the amplitude approaches zero. Second, the satellite-by-satellite phase changes introduced by configuration changes should be predictable or measurable, and the uncertainty of the phase propagation residuals should not exceed the allowable range. The phase fidelity index evaluates the degree to which the phase relationship is preserved by space processing or configuration switching, and is not the same as the carrier phase variance, which simply describes the accuracy of random phase measurements.

[0095] Furthermore, the candidate configuration can satisfy at least one of the following constraints: lower limit of satellite response amplitude, upper limit of phase change uncertainty, upper limit of weight norm, radio frequency dynamic range, interference suppression degree, and phase transfer availability. The space processing configuration can be obtained through candidate library traversal, convex optimization, sequential quadratic programming, or manifold optimization; when discrete physical actions are involved, mixed integer search or finite candidate enumeration can be used. The above algorithms are merely means to obtain the target configuration and are not limited to necessary steps of this invention.

[0096] See Figure 5When it is necessary to switch from the current space processing configuration to the target space processing configuration (i.e., the target configuration), the old and new processing branches can be parallelized for a short period of time based on the same batch of multi-element synchronous samples to measure the phase transfer and uncertainty of each satellite. If the measurement results of the phase transfer and its uncertainty meet the preset threshold requirements, weight replacement and phase compensation are submitted simultaneously, and the switch to the target space processing configuration is performed; if the preset threshold requirements are not met, the switch to the target space processing configuration is not performed this time, and the current space processing configuration continues to be used.

[0097] For changes in physical observation configuration, geometric phase changes can be predicted before physical actions are performed, and compensation can be made based on attitude or displacement measurements after the actions are performed. The residual uncertainty after compensation is included in the observation noise covariance matrix. For target configurations that fail the corresponding switching condition test, the corresponding configuration switch will not be performed this time, and the candidate configurations will be re-evaluated. After the target configuration that meets the preset switching conditions is re-determined, the subsequent step S107 will be performed based on the target configuration.

[0098] In step S107, based on the target configuration, floating-point estimation, integer least squares estimation, and integrity check are performed to obtain the navigation state solution results.

[0099] This step acquires or forms corresponding observations based on the target configuration determined in step S106, and performs floating-point estimation using the corresponding joint observation model and observation noise covariance. The floating-point estimation temporarily treats the integer ambiguity as a real number, and solves it together with the continuous navigation state and the continuous interference parameters that need to be explicitly estimated, to obtain the floating-point ambiguity, the continuous navigation state, and its covariance.

[0100] In some implementations, the floating-point ambiguity is updated using spatial manifold observations and covariance terms and cross-correlation terms in the observation noise covariance matrix corresponding to the spatial manifold observations, so that the correlation of co-origin observations is included in the floating-point estimation weights.

[0101] Subsequently, integer least squares estimation is performed, searching for integer candidates within the integer domain that minimize the weighted residual or equivalent objective. This integer least squares estimation can typically employ a conventional integer search. In some implementations, it can also construct integer candidate scoring terms or reorder integer candidates based on updated floating-point ambiguities using spatial manifold observations. Thus, floating-point updates and integer candidate processing correspond to two consecutively executed estimation stages. These implementations are termed "integer least squares estimation based on spatial manifold observation constraints," meaning that spatial manifold observations participate in integer candidate evaluation through a joint model, rather than directly setting the spatial manifold as an integer constraint. Compared to conventional integer least squares estimation based solely on carrier floating-point solutions and covariance, it adds spatial response consistency evaluation; simultaneously, compared to integer estimation using known baseline length as a geometric constraint, the constraint source is the spatial manifold and its statistical model between arrays or configurations.

[0102] For example, for each integer candidate, the corresponding continuous navigation state and continuous interference parameters can be recovered. Applicable terms from the carrier residual, code residual, coherent synthesis residual, and spatial manifold residual are calculated, and a candidate score is formed using the joint covariance. In this way, candidates with similar traditional carrier residuals but inconsistent predicted spatial responses can be further distinguished. Subsequently, a fixed solution is obtained from the integer candidates, and the target navigation state in position, velocity, attitude, baseline, or clock bias is output.

[0103] Integrity checks are used to evaluate the reliability of integer fixed results and navigation state solution results. For example... Figure 6 As shown, the integrity check can output at least one of the following: coherent sensitivity gain, spatial manifold condition information gain, correlation between the two types of observations, minimum discrimination distance, candidate ratio or residual test value, correct fixation probability estimate, and incorrect fixation risk estimate. It can further combine the number of satellites, geometric conditions, and residual consistency for a comprehensive evaluation to determine whether to accept the fixed solution. Among these, coherent sensitivity gain, spatial manifold condition information gain, and correlation between the two types of observations are used to assist in characterizing the observation and model quality, and are not sufficient conditions for accepting the fixed solution on their own. Whether to accept the fixed solution can also be determined by combining at least one of the following: candidate ratio, residual test value, correct fixation probability estimate, incorrect fixation risk estimate, number of satellites, geometric conditions, and residual consistency. When the comprehensive evaluation result meets the preset integrity requirements, the fixed solution is accepted and the corresponding navigation state solution result is output; otherwise, the fixed solution is not accepted.

[0104] To better aid in understanding the method provided by this invention, four specific embodiments are provided below: Example 1: Interference-free four-element precision measurement This embodiment is designed for open, unobstructed, and interference-free short-baseline precision measurement scenarios, such as slope or bridge displacement monitoring and short-baseline calculation between reference stations. The receiver employs a four-element planar array or L-shaped array, with an element spacing of 0.48 times the carrier wavelength in the example. At the L1 frequency of 1575.42MHz, the carrier wavelength is approximately 19.03 cm, corresponding to an element spacing of approximately 9.1 cm. The four RF channels share a common reference clock and sampling clock. After power-on, an initial element sample buffer is established, and the element noise covariance is estimated in idle frequency bands or when there is no signal.

[0105] In the coherent composite observation generation process of step S102, common weights that satisfy the multi-satellite response consistency constraint are used to ensure that the composite complex gain corresponding to different satellites remains consistent, approximately consistent, or can be determined through calibration, thereby helping to maintain the cancellability of inter-satellite common-mode phase. The weights can improve the carrier-to-noise ratio (CNR) of the lowest-response satellite under the example constraint that the response amplitude of each satellite is not less than -3 dB of the relative peak value. In an exemplary evaluation, the effective CNR is improved by approximately 4.5 dB to 5.5 dB, and the high-elevation-angle satellite's CNR is improved from approximately 45 dB-Hz to approximately 50 dB-Hz. When the CNR is improved by 5 dB, the carrier phase variance is reduced to approximately 1 / 3.2 of its original value, and the standard deviation decreases by approximately 44%.

[0106] During the generation of spatial manifold observations, using the first element as a reference, complex ratios are constructed between the other three elements and the first element for each satellite. Each complex ratio can be decomposed into the amplitude ratio and phase difference between elements, or it can be used as a whole as complex projective coordinates after eliminating common complex gain. The three complex ratios are equivalent to six real observations, retaining M minus 1 complex spatial degrees of freedom. If nonlinear complex ratios are used, their covariance is propagated based on the current response and Jacobian. Coherent composite observations and complex ratios are formed from the same batch of samples, thus the joint covariance matrix retains the cross-correlation terms between the two.

[0107] The joint model may include double-difference carrier observations, code observations, coherent synthetic observations, and spatial manifold observations, as well as baseline or attitude, double-difference integer ambiguities, common complex gain, channel amplitude and phase parameters, phase center deviation, and residual clock bias. Valid conditional information for integer ambiguities is determined through reference constraints, credible calibration priors, and Schur complement.

[0108] To differentiate between spatial manifold information and carrier-to-noise ratio (CNR) gain, an equal output CNR comparison can be established: the first group uses only the coherent synthetic observation generation process, while the second group uses both the coherent synthetic observation generation process and the spatial manifold observation generation process simultaneously at the same output CNR, taking into account their cross-correlation. In an exemplary result, the second group's integer ambiguity effective conditional information matrix... The minimum eigenvalue was increased by approximately 1.8 dB to 2.6 dB, the initial fixation time was shortened from approximately 60 seconds to approximately 35 seconds, and the fixation success rate increased from 96.2% to 99.1%. The fixation success rate can be determined by the proportion of correct integer fixations in repeated trials or statistical epochs to the total number of fixations that met the fixation criterion.

[0109] When performing the above-mentioned output signal-to-noise ratio comparison, it is advisable to maintain consistency in the number and geometry of satellites, baseline length, observation duration, noise model, fixed integer criteria, and number of statistical samples, and to record the output signal-to-noise ratio separately. The minimum eigenvalue, the first fixed time, and the fixed result.

[0110] Example 2: Selection of Dangerous Integer Direction-Driven Weights This embodiment is designed for integrated positioning and orientation scenarios with weak to moderate interference or requiring rapid integer fixing. First, an offline candidate weight library is generated, containing different null orientations, null depths, whitening degrees, and multi-satellite trade-offs, with an example of 16 to 64 candidates. Alternatively, weights that meet the constraints can be generated online based on the current interference covariance, and these weights, together with the current weights, constitute the candidate space processing configuration.

[0111] Based on the floating-point ambiguity corresponding to the current configuration or the configuration of the previous epoch, or based on the initial floating-point estimate of the current epoch, an integer least squares candidate search is performed to obtain the best integer candidate and the second-best integer candidate. The difference vector between the two, along with the integer Voronoi neighbors, short integer vectors, and integer directions corresponding to historical error fixed events, are used to form a set of dangerous integer directions. During online computation, 10 to 50 dangerous integer directions can be retained from the set of dangerous integer directions in ascending order of discrimination distance. For each candidate weight, the joint covariance of coherent synthetic observations and spatial manifold observations is updated, the effective conditional information after eliminating continuous interference parameters is calculated, and the minimum discrimination distance is obtained.

[0112] Configuration selection considers carrier-to-noise ratio gain, minimum discrimination distance, and handover cost simultaneously, and can impose example constraints such as phase change uncertainty not exceeding 0.05 cycles, interference suppression not less than 25 dB, and satellite response amplitude and RF dynamic range meeting limits. Target configurations can be determined through candidate traversal, constraint optimization, or hierarchical screening.

[0113] To maintain carrier phase continuity, the old and new weights can be run in parallel for 0.2 to 1 second using the same dual-branch system before the switchover, measuring the satellite-by-satellite phase transfer and its standard deviation. Once the threshold is met, weight replacement and phase compensation are submitted synchronously. The ratio test value for integer candidates can be defined as the weighted residual target value of the suboptimal integer candidates. Weighted residual objective value of the best integer candidate The ratio, that is ,in Not less than The larger the ratio, the higher the degree of separation between the two candidates.

[0114] In one exemplary result, the minimum discrimination distance improved by approximately 2.4 dB, the ratio check value increased from 1.15 to 1.90, the error fixation rate decreased from 1.2% to 0.15%, and no cycle slip occurred during the switching process. The dB improvement in minimum discrimination distance can be adjusted based on the target configuration and the current configuration. The ratio is usually determined by taking 10 times the logarithm; whether a cycle slip occurs can be judged by combining the receiver lock indication and the continuity of the phase residual.

[0115] Example 3: Correlation Processing of Multiple Outputs from the Same Source In this embodiment, the receiver simultaneously generates one coherent synthesized output, two subarray outputs, three element ratio outputs, and one adaptive null output. All outputs are generated from the same multi-element synchronous samples through different processing matrices. The output dimension may be greater than the number of elements, and therefore the joint covariance matrix may be singular. When the multi-processing vectors are approximately collinear, the joint covariance matrix may be nearly singular.

[0116] The joint covariance matrix is ​​eigenvalued, and the effective rank is determined according to a relative threshold. Whitened observations are then formed using the effective eigenspace, or a pseudo-inverse is used in information computation. For example, the eigenvalues ​​of the joint covariance matrix are arranged in descending order, and the effective rank threshold is reached when the ratio of the i-th eigenvalue to the largest eigenvalue is not lower than the effective rank threshold. At the same time, retain the corresponding feature direction; The value is determined based on the noise floor, sample size, and numerical precision. The resulting information rank does not exceed the independent dimension supported by the original samples. As an equivalent implementation, the original array element samples can also be directly retained as sufficient statistics, and a corresponding model can be built in the original sample domain.

[0117] Furthermore, the system calculates the marginal information increment path by path: removing one output from the joint observations and recalculating the effective conditional information and minimum discrimination distance. Under the condition of using the same set of dangerous integer directions and the same parameters to be conditionalized, the ratio of the minimum discrimination distance corresponding to all observations to the minimum discrimination distance obtained after removing the corresponding observation can be taken as 10 times the common logarithm, which is used as the marginal discrimination information increment corresponding to that observation. If this change is lower than the example threshold of 0.2 dB, the corresponding processing path can be shut down to reduce computational load and power consumption; if an output can improve the discrimination of dangerous integer directions, it can be retained even if its single-path signal-to-noise ratio is low.

[0118] If new physical observations are formed through changes in array attitude, observation time, or frequency, the system does not determine their independence based on the number of digital outputs. Instead, it updates the joint covariance based on the actual temporal, spatial, and frequency correlations. This distinguishes between deterministic multiple outputs of the same sample and new observations under different physical states.

[0119] Example 4: Active Physics Observation Configuration This embodiment applies to vehicle-mounted or airborne platforms. The platform can change the array attitude via a turntable or carrier, switching between multiple physical subarrays, or switching between or simultaneously activating L1 and L5 frequency points. An exemplary attitude candidate range is ±15 degrees for azimuth and ±10 degrees for pitch. Each attitude, physical subarray, or frequency point combination forms a candidate physical observation configuration, and its statistical correlation is determined according to the actual temporal and spatial relationships.

[0120] Before executing a physical action, the system configures the predicted satellite complex response, output signal-to-noise ratio, joint covariance, effective condition information, and minimum discrimination distance for each candidate physical observation. The geometric phase change introduced by the physical action can be expressed as... ,in, This indicates the predicted geometric phase change. λ represents the path change of the array element phase center along the satellite line of sight, and λ represents the carrier wavelength at the corresponding frequency point; when When the uncertainty exceeds a threshold, the corresponding configuration is not selected. For example, the minimum discrimination distance improvement can be calculated based on the difference between the time before and after the action. The ratio is usually determined by the logarithm, which is 10 times. Physical actions are only performed when the improvement exceeds 1.5 dB, the phase continuity meets the requirements, and the overall index still improves after deducting the switching cost.

[0121] During the operation, affected observations can be marked as unavailable or their variance increased accordingly. After the operation is completed, a new attitude is determined based on the orientation results of inertial devices or dual antennas, compensating for predictable geometric phase changes and incorporating residual uncertainties into the observation noise. When switching or adding L5, the integer ambiguity of the newly added frequency points is reinitialized, and the discrimination distance of some dangerous integer directions can be improved by using dual-frequency wide-lane combinations.

[0122] After completing the actions and phase compensation, the system uses the target physical observation configuration to perform floating-point estimation, integer least squares estimation, and integrity verification. If the actual response after the actions deviates significantly from the predicted response, the floating-point state can be maintained and the physical state or continuous disturbance parameters can be re-estimated; this processing is an optional implementation method and does not change the basic process of navigation calculation based on the target configuration.

[0123] Based on the same inventive concept, embodiments of the present invention also provide an array-guided dual-gain precision observation device, comprising: The sample acquisition module is used to acquire multi-element synchronization samples and their noise or interference statistics from the satellite navigation receiver. The observation generation module is used to generate coherent composite observations and spatial manifold observations based on multi-element synchronous samples; wherein, the coherent composite observations are used to improve the carrier-to-noise ratio, and the spatial manifold observations preserve the spatial response differences between different array elements, different subarrays, different spatial configurations, or different physical observation states; The covariance construction module is used to construct a joint covariance matrix based on noise or interference statistics, including cross-correlation terms of coherent synthetic observations and spatial manifold observations generated from the same multi-element synchronous samples. The joint modeling module is used to select at least two of the following: carrier observation, code observation, coherent synthesis observation, and spatial manifold observation, with at least one of them being coherent synthesis observation or spatial manifold observation, to construct a joint observation model that includes continuous navigation state, integer ambiguity, and continuous interference parameters. The information determination module is used to select different spatial processing configurations or physical observation configurations as candidate configurations, and for each candidate configuration, to determine the observation noise covariance matrix corresponding to the selected observation from the joint covariance matrix. Combined with the joint observation model, the continuous interference parameters are eliminated by projection or Schur complement to obtain effective condition information corresponding to each candidate configuration and related to the integer ambiguity. The configuration selection module is used to calculate the discrimination distance of each candidate configuration in multiple dangerous integer directions using effective condition information, and to determine the target configuration from the candidate configurations based on at least two of the following: output signal-to-noise ratio, phase fidelity index, satellite response amplitude, and minimum discrimination distance. The navigation solution module is used to perform floating-point estimation, integer least squares estimation and integrity checks based on the target configuration to obtain the navigation state solution results.

[0124] Each module in the above-mentioned device can be implemented by a processor executing program instructions stored in memory, or by software components, hardware circuits, or a combination of both. The module name is only used to distinguish its main function and does not require that each module correspond to a separate physical device; a processing unit can implement the functions of multiple modules, and the same module can also be distributed on multiple controlled computing nodes.

[0125] Based on the same inventive concept, embodiments of the present invention also provide a storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the array navigation dual-gain precision observation method as provided in any embodiment of the present invention.

[0126] In embodiments of the present invention, any combination of one or more storage media may be used. The storage medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, RAM, ROM, erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in connection with an instruction execution system, apparatus, or device.

[0127] Based on the same inventive concept, embodiments of the present invention also provide an electronic device. Figure 7 The diagram shown illustrates the structure of the electronic device 100 provided in an embodiment of the present invention. In some embodiments, the electronic device may be a mobile phone, tablet computer, wearable device, in-vehicle device, augmented reality (AR) / virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), or other terminal device. Furthermore, the array navigation dual-gain precision observation method provided in this embodiment can also be applied to databases, servers, and service response systems based on terminal artificial intelligence. This embodiment does not impose any limitations on the specific application scenarios of the array navigation dual-gain precision observation method.

[0128] like Figure 7 As shown, the electronic device 100 provided in this embodiment of the invention includes a memory 101 and a processor 102.

[0129] The memory 101 is used to store computer programs; preferably, the memory 101 includes various media that can store program code, such as ROM, RAM, magnetic disk, USB flash drive, memory card or optical disk.

[0130] Specifically, memory 101 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory. Electronic device 100 may further include other removable / non-removable, volatile / non-volatile computer system storage media. Memory 101 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0131] The processor 102 is connected to the memory 101 and is used to execute the computer program stored in the memory 101 to implement the array navigation dual-gain precision observation method provided in any embodiment of the present invention.

[0132] In an optional implementation, the processor 102 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, evaluation index gate or transistor logic devices, or evaluation index hardware components.

[0133] In an optional embodiment, the electronic device 100 of this invention may further include a display 103. The display 103 is communicatively connected to the memory 101 and the processor 102, and is used to display the relevant GUI interactive interface of the array navigation dual-gain precision observation method.

[0134] It is understood that the array navigation dual-gain precision observation device, storage medium and electronic device provided in the embodiments of the present invention correspond to the array navigation dual-gain precision observation method provided in the embodiments of the present invention. The explanation, examples and beneficial effects of the relevant contents can be referred to the corresponding parts of the method, and will not be repeated here.

[0135] In summary, compared with existing technologies, it has the following beneficial effects: 1. Coherent synthetic observations and spatial manifold observations are generated based on synchronous samples from the same batch of multi-element arrays. Coherent synthetic observations improve the output signal-to-noise ratio or carrier-to-noise ratio, which can be reflected in the reduction of carrier phase variance, the reduction of cycle slip risk caused by noise, and the reduction of floating-point ambiguity covariance; spatial manifold observations preserve the spatial response differences between different array elements, subarrays, spatial configurations, or physical observation states, enabling them to participate in navigation state estimation and integer candidate discrimination, and achieving the synergistic utilization of sensitivity gain and conditional information gain.

[0136] 2. The shared noise of the same source observations is explicitly processed by the cross-correlation terms in the joint covariance matrix, and the singular or near-singular statistical structures are processed by effective rank whitening, pseudo-inverse or redundant observation reduction; further, the continuous interference parameters are eliminated by projection or Shur complement to obtain effective conditional information related to integer ambiguity, so that the available information is constrained by the effective rank of the original sample, and the overestimation of the discrimination ability caused by repeated inclusion of relevant information or neglect of parameter coupling is reduced.

[0137] 3. Utilize effective condition information to calculate the discrimination distance of each candidate configuration in multiple dangerous integer directions, and comprehensively evaluate candidate configurations by outputting signal-to-noise ratio, phase fidelity index, satellite response amplitude, and minimum discrimination distance. This ensures that configuration selection balances reception quality with the degree of separation in easily confused integer directions. Further consideration of phase change uncertainty, weight norm, RF dynamic range, and interference suppression can balance integer discrimination capability, carrier observation availability, and configuration implementation feasibility.

[0138] 4. Based on the target configuration, floating-point estimation, integer least squares estimation, and integrity checks are performed. Spatial manifold observations and their covariance are used to update floating-point ambiguities and evaluate or rearrange integer candidates, thereby further distinguishing candidates with similar carrier residuals but inconsistent spatial responses. Incorporating sensitivity gain, spatial manifold conditional information gain, observation correlation, minimum discrimination distance, and fixed risk into the integrity evaluation can improve the reliability of integer fixed results and navigation state calculation results.

[0139] 5. For deterministic multiple outputs formed by the same sample, sufficient statistics or non-redundant outputs of the original array elements are retained to maintain the statistical boundaries of effective information; for physical observation configurations formed by changes in array attitude, physical subarrays, frequency points, time or trajectory states, the differences in new observations are evaluated based on actual spatiotemporal correlation, making the present invention applicable to scenarios such as interference-free precision measurement, interference of different intensities, displacement monitoring, and integrated positioning and orientation reception.

[0140] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0141] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dual-gain precision observation method for array navigation, characterized in that, include: Acquire multi-element synchronization samples and their noise or interference statistics from a satellite navigation receiver; Based on multi-element synchronous samples, coherent synthetic observations and spatial manifold observations are generated; wherein, the coherent synthetic observations are used to improve the carrier-to-noise ratio, and the spatial manifold observations preserve the spatial response differences between different array elements, different subarrays, different spatial configurations, or different physical observation states. Based on noise or interference statistics, a joint covariance matrix is ​​constructed, which includes cross-correlation terms of coherent synthetic observations and spatial manifold observations generated from the same multi-element synchronous samples. Select at least two of the following: carrier observation, code observation, coherent synthesis observation, and spatial manifold observation, and at least one of them must be coherent synthesis observation or spatial manifold observation, and construct a joint observation model that includes continuous navigation state, integer ambiguity, and continuous interference parameters. Different spatial processing configurations or physical observation configurations are used as candidate configurations. For each candidate configuration, the observation noise covariance matrix corresponding to the selected observation is determined from the joint covariance matrix. Combined with the joint observation model, the continuous interference parameters are eliminated by projection or Schur complement to obtain effective condition information corresponding to each candidate configuration and related to the integer ambiguity. Using valid condition information, calculate the discrimination distance of each candidate configuration in multiple dangerous integer directions, and determine the target configuration from the candidate configurations based on at least two of the following: output signal-to-noise ratio, phase fidelity index, satellite response amplitude, and minimum discrimination distance; Based on the target configuration, floating-point estimation, integer least squares estimation, and integrity checks are performed to obtain the navigation state solution results.

2. The array navigation dual-gain precision observation method as described in claim 1, characterized in that, The generation process of the coherent composite observations includes: The coherent synthesis weights are determined based on at least one of the following criteria: maximum signal-to-noise ratio criterion, minimum variance distortionless response criterion, linear constraint criterion, multi-star common weight constraint, and interference suppression constraint. Based on the coherent synthesis weights, the multi-element synchronous samples are coherently synthesized to generate the coherent synthesized observations.

3. The array navigation dual-gain precision observation method as described in claim 1, characterized in that, The generation process of the spatial manifold observations includes: Based on the multi-element synchronization samples, at least one of the following is obtained: subarray output, amplitude ratio between elements, phase difference between elements, complex projective coordinates after eliminating common complex gain, and phase transfer between different spatial processing configurations, and is determined as the spatial manifold observation.

4. The array navigation dual-gain precision observation method as described in claim 1, characterized in that, Constructing the joint covariance matrix includes: When the coherent composite observation and the spatial manifold observation are generated from the same multi-element synchronous sample, the coherent composite weights used to generate the coherent composite observation and the spatial manifold processing matrix used to generate the spatial manifold observation are determined. Based on the original array element noise or interference covariance matrix, the coherent synthesis weights, and the spatial manifold processing matrix, the covariance of the coherent synthesis observation, the covariance of the spatial manifold observation, and the cross covariance of the two are determined, and the joint covariance matrix is ​​formed accordingly.

5. The array navigation dual-gain precision observation method as described in claim 1, characterized in that, The continuous interference parameters include at least one of the following: common complex gain, channel amplitude and phase parameters, multipath parameters, interference direction, and interference intensity. The process of eliminating the continuous interference parameters includes: Based on the observation noise covariance matrix and the joint observation model, determine the tangent space or information matrix block corresponding to the continuous interference parameters; The joint observation model is projected onto the orthogonal complement space of the tangent space corresponding to the continuous interference parameters, or the Schur complement is obtained by dividing the information matrix corresponding to the continuous interference parameters into elimination blocks, so as to obtain effective conditional information related to the integer ambiguity.

6. The array navigation dual-gain precision observation method as described in claim 1, characterized in that, The process of determining the multiple dangerous integer directions includes: The plurality of dangerous integer directions are determined based on at least one of integer least squares candidate differences, integer Voronoi neighbors, short integer vector sets, and historical error fixation events.

7. The array navigation dual-gain precision observation method as described in claim 1, characterized in that, Performing the floating-point estimation includes: The floating-point ambiguity is updated using the spatial manifold observations and the covariance and cross-correlation terms in the observation noise covariance matrix corresponding to the spatial manifold observations. Performing the integer least squares estimation includes: Based on the updated floating-point ambiguity, integer candidate scoring items are constructed or integer candidates are reordered using the spatial manifold observations.

8. A dual-gain array-guided precision observation device, characterized in that, include: The sample acquisition module is used to acquire multi-element synchronization samples and their noise or interference statistics from the satellite navigation receiver. The observation generation module is used to generate coherent composite observations and spatial manifold observations based on multi-element synchronous samples; wherein, the coherent composite observations are used to improve the carrier-to-noise ratio, and the spatial manifold observations preserve the spatial response differences between different array elements, different subarrays, different spatial configurations, or different physical observation states; The covariance construction module is used to construct a joint covariance matrix based on noise or interference statistics, including cross-correlation terms of coherent synthetic observations and spatial manifold observations generated from the same multi-element synchronous samples. The joint modeling module is used to select at least two of the following: carrier observation, code observation, coherent synthesis observation, and spatial manifold observation, with at least one of them being coherent synthesis observation or spatial manifold observation, to construct a joint observation model that includes continuous navigation state, integer ambiguity, and continuous interference parameters. The information determination module is used to select different spatial processing configurations or physical observation configurations as candidate configurations, and for each candidate configuration, to determine the observation noise covariance matrix corresponding to the selected observation from the joint covariance matrix. Combined with the joint observation model, the continuous interference parameters are eliminated by projection or Schur complement to obtain effective condition information corresponding to each candidate configuration and related to the integer ambiguity. The configuration selection module is used to calculate the discrimination distance of each candidate configuration in multiple dangerous integer directions using effective condition information, and to determine the target configuration from the candidate configurations based on at least two of the following: output signal-to-noise ratio, phase fidelity index, satellite response amplitude, and minimum discrimination distance. The navigation solution module is used to perform floating-point estimation, integer least squares estimation and integrity checks based on the target configuration to obtain the navigation state solution results.

9. A storage medium, characterized in that, It stores a computer program, wherein when the computer program is executed by a processor, it implements the array navigation dual-gain precision observation method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, The electronic device includes: A processor, a memory, and program instructions stored in the memory and executable by the processor, wherein the processor, when executing the program instructions, implements the array navigation dual-gain precision observation method as described in any one of claims 1 to 7.

Images